'''This is essentially the groups meeting notes – a trail of bread crumbs of topics of conversation and projects entertained by the group'''
(note: Learn more about previous neuro research at Noisebridge on the wiki... For example, the [[Analog_EEG_Amp]] page describes some project ideas and work done by others here in 2012)
===Websites and events that have piqued our interest===
https://noisebridge.net/wiki/NBDSM -- noiseBridge Deepnet and Statistical Mechanics -- first meetup @ noisebridge 7/6/17 at 7PM

https://metacademy.org/
-- machine learning knowledge graph

https://machinelearningguide.libsyn.com/rss -- machine learning guide podcast

http://www.thetalkingmachines.com/ -- podcast

https://karpathy.github.io/2015/05/21/rnn-effectiveness/

http://alexandre.barachant.org/papers/

http://ncs.ethz.ch/publications -- neuromorphic cognitive systems

https://github.com/crillab/gophersat/blob/master/examples/sat-for-noobs.md -- SAT solvers

https://media.ccc.de/v/34c3-8948-low_cost_non-invasive_biomedical_imaging -- Open EIT 34c3 talk https://github.com/OpenEIT

http://acrovirt.org/ -- sensors

http://www.neuroeducate.com/ -- citizen neuroscience

https://www.youtube.com/watch?v=9mZuyUzyN4Q -- "Categories for the Working Hacker"

http://radicalsciencenews.org/599-2/ -- "Deep Learning Fuels Nvidia’s Self-Driving Car Technology"

== EEG Electrodes ==

https://sites.google.com/site/biofeedbackpages/velcro-sensors -- Saline electrodes

== Generative Adversarial Networks (GAN) ==

https://arxiv.org/pdf/1710.08864 -- "One pixel attack for fooling deep neural networks"

== OpenCV ==

http://arnab.org/blog/so-i-suck-24-automating-card-games-using-opencv-and-python

http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.185.585&rep=rep1&type=pdf -- "Pyramidal Implementation of the Lucas Kanade Feature Tracker Description of the algorithm"

== Category Theory ==

https://arxiv.org/pdf/1711.10455 -- "Backprop as Functor: A compositional perspective on supervised learning"

http://math.ucr.edu/home/baez/rosetta.pdf -- "Physics, Topology, Logic and Computation: A Rosetta Stone"

https://www.youtube.com/watch?v=BF6kHD1DAeU -- "Category theory foundations 1.0 — Steve Awodey"

== Proof Searcher ==

https://arxiv.org/pdf/cs/0207097 -- "Optimal Ordered Problem Solver"

http://people.idsia.ch/~juergen/ultimatecognition.pdf -- "Ultimate Cognition a la Gödel"

http://people.idsia.ch/~juergen/selfreflection.pdf -- "Towards an Actual Gödel Machine Implementation"

== Capsule Models ==

https://arxiv.org/pdf/1710.09829.pdf -- "Dynamic Routing Between Capsules"

https://openreview.net/pdf?id=HJWLfGWRb -- "Matrix Capsules with EM Routing"

== Infrared Neuroimaging ==

http://www.ecse.rpi.edu/~yazici/bio_book.pdf -- "Near Infrared Imaging and Spectroscopy for Brain Activity Monitoring"

== Geometry ==

http://arxiv.org/abs/1710.10784 -- "How deep learning works --The geometry of deep learning"

== Affective Computing ==

http://affect.media.mit.edu/pdfs/05.ahn-picard-acii.pdf -- "Affective-Cognitive Learning and Decision
Making: A Motivational Reward Framework For Affective Agents"

== Explainability ==

http://arxiv.org/abs/1708.01785 -- "Interpreting CNN knowledge via an Explanatory Graph"

== NLP ==

https://arxiv.org/pdf/1605.06640 -- "Programming with a Differentiable Forth Interpreter"

https://pdfs.semanticscholar.org/f683/dbe8a22d633ad3a2cff379b055b26684a838.pdf -- "Solving General Arithmetic Word Problems"

https://arxiv.org/pdf/1611.04558.pdf -- "Google’s Multilingual Neural Machine Translation System: Enabling Zero-Shot Translation"

http://emnlp2014.org/papers/pdf/EMNLP2014162.pdf -- "GloVe: Global Vectors for Word Representation"

== RNNs ==

https://arxiv.org/pdf/1611.01576.pdf -- "Quasi Recurrent Neural Networks"

== Hyper-parameter Optimization ==

https://arxiv.org/abs/1603.06560 -- "Hyperband: A Novel Bandit-Based Approach to Hyperparameter Optimization"

== Transfer Learning ==

http://arxiv.org/abs/1710.10776v1 -- "Transfer Learning to Learn with Multitask Neural Model Search"

== Reinforcement Learning ==

http://www2.hawaii.edu/~sstill/StillPrecup2011.pdf -- "An information-theoretic approach to curiosity-driven reinforcement learning"

https://arxiv.org/abs/1605.06676 -- "Learning to Communicate with Deep Multi-Agent Reinforcement Learning"

== Learning to Learn ==

https://arxiv.org/pdf/1703.01041.pdf -- "Large-Scale Evolution of Image Classifiers"

https://arxiv.org/pdf/1611.01578 -- "Neural Architecture Search with Reinforcement Learning"

== The Utility of "Noise" in ML ==

https://www.cs.toronto.edu/~hinton/absps/JMLRdropout.pdf -- "Dropout: A Simple Way to Prevent Neural Networks from Overfitting"

http://yann.lecun.com/exdb/publis/pdf/lecun-90b.pdf -- "Optimal Brain Damage"

https://arxiv.org/pdf/1502.01852.pdf -- "Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification"

== One-shot learning ==

https://arxiv.org/abs/1605%2E06065 -- "One-shot Learning with Memory-Augmented Neural Networks"

== Program Synthesis ==

https://pdfs.semanticscholar.org/0163/35ce7e0a073623e1deac7138b28913dbf594.pdf -- "Human-level concept learning through probabilistic program induction"

https://arxiv.org/pdf/1511.06279.pdf -- "Neural Programmer: Inducing Latent Programs with Gradient Descent"

https://arxiv.org/abs/1608.04428 -- "TerpreT: A Probabilistic Programming Language for Program Induction" Gaunt et al 2016

== Machine Learning Interaction ==

https://teachablemachine.withgoogle.com/#

== Game Theory ==

https://arxiv.org/abs/1707.01068v1 - Maintaining cooperation in complex social dilemmas using deep reinforcement learning

== Questions of Physics and Free Will ==

http://www.scottaaronson.com/papers/giqtm3.pdf - The Ghost in the Quantum Turing Machine

== CNN ==

https://adeshpande3.github.io/adeshpande3.github.io/A-Beginner's-Guide-To-Understanding-Convolutional-Neural-Networks/ - "A Beginner's Guide To Understanding Convolutional Neural Networks"

https://adeshpande3.github.io/A-Beginner's-Guide-To-Understanding-Convolutional-Neural-Networks-Part-2/ - "A Beginner's Guide To Understanding Convolutional Neural Networks Part 2"

http://scs.ryerson.ca/~aharley/vis/harley_vis_isvc15.pdf -- "An Interactive Node-Link Visualization
of Convolutional Neural Networks"

http://white.stanford.edu/teach/index.php/An_Introduction_to_Convolutional_Neural_Networks

http://www.cv-foundation.org/openaccess/content_cvpr_2015/papers/Dosovitskiy_Learning_to_Generate_2015_CVPR_paper.pdf -- "Learning to Generate Chairs With Convolutional Neural Networks"

http://www.pamitc.org/cvpr15/files/lecun-20150610-cvpr-keynote.pdf
-- "What's Wrong With Deep Learning?"

== Mind-Body Relations ==

http://www.pnas.org/content/111/20/7379.full.pdf -- "Voluntary activation of the sympathetic nervous system and attenuation of the innate immune response in humans"

== Math ==

https://arxiv.org/pdf/1311.1090.pdf -- "Polyhedrons and Perceptrons Are Functionally Equivalent"

Example code and training data using polyhedrons developed by author of above paper: https://www.noisebridge.net/wiki/DreamTeam#Code

== Bayesian Inference ==

https://noisebridge.net/images/e/ef/Perception_is_in_the_Details12.pdf --
"Perception is in the Details: A Predictive Coding Account of the Psychedelic Phenomenon"

http://rsif.royalsocietypublishing.org/content/10/86/20130475 --
"Life as we know it"

http://jmlr.csail.mit.edu/proceedings/papers/v31/wang13b.pdf --
"Collapsed Variational Bayesian Inference for Hidden Markov Models"

http://www.datalab.uci.edu/papers/nips06_cvb.pdf --
"A Collapsed Variational Bayesian Inference Algorithm for Latent Dirichlet Allocation"

http://www.cnbc.cmu.edu/~tai/papers/lee_mumford_josa.pdf --
"Hierarchical Bayesian inference in the visual cortex"

https://www.researchgate.net/profile/Til_Bergmann/publication/262423308_Temporal_coding_organized_by_coupled_alpha_and_gamma_oscillations_prioritize_visual_processing/links/0deec537d1bfda474c000000/Temporal-coding-organized-by-coupled-alpha-and-gamma-oscillations-prioritize-visual-processing.pdf --
"Temporal coding organized by coupled alpha and gamma oscillations prioritize visual processing"

http://www.cell.com/neuron/pdf/S0896-6273(15)00823-5.pdf --
"Rhythms for Cognition: Communication through Coherence"

http://www.biorxiv.org/content/biorxiv/early/2014/05/06/004804.full.pdf --
"Visual Areas Exert Feedforward and Feedback Influences through Distinct Frequency Channels"

== Speech Recognition ==

https://arxiv.org/pdf/1612.00694v1 -- "ESE: Efficient Speech Recognition Engine with Compressed LSTM on FPGA"

== Sound Classification ==

https://arxiv.org/pdf/1608.04363v2 -- "Deep Convolutional Neural Networks and Data Augmentation for Environmental Sound Classification"

https://arxiv.org/pdf/1605.09507 "Deep convolutional neural networks for predominant instrument recognition in polyphonic music"

== Hardware Implementations - FPGA, GPU, etc ==

https://www.cse.iitk.ac.in/users/isaha/Publications/Journals/NC10.pdf --
"Artificial neural networks in hardware: A survey of two decades of progress"

http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.133.9185&rep=rep1&type=pdf
"A Self-Repairing Multiplexer-Based FPGA Inspired by Biological Processes"

http://www.genetic-programming.com/jkpdf/fpga1998.pdf -- "Evolving Computer Programs using Rapidly Reconfigurable Field-Programmable Gate Arrays and Genetic Programming"

http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.114.2588&rep=rep1&type=pdf -- "Flexible Implementation of Genetic Algorithms on FPGAs"

http://www.users.muohio.edu/jamiespa/html_papers/gem_10.pdf -- "Revisiting Genetic Algorithms for the FPGA Placement Problem"

https://arxiv.org/pdf/1609.09296v1 -- "Comprehensive Evaluation of OpenCL-based Convolutional Neural Network Accelerators in Xilinx and Altera FPGAs"

http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.409.7533&rep=rep1&type=pdf -- "FPGA-TARGETED NEURAL ARCHITECTURE FOR EMBEDDED ALERTNESS DETECTION"

https://arxiv.org/pdf/1611.02450v1 -- "PipeCNN: An OpenCL-Based FPGA Accelerator for Large-Scale Convolution Neuron Networks"

https://arxiv.org/pdf/1605.06402v1 -- "Ristretto: Hardware-Oriented Approximation of Convolutional Neural Networks"

https://homes.cs.washington.edu/~luisceze/publications/snnap-hpca-2015.pdf -- "SNNAP: Approximate Computing on Programmable SoCs via Neural Acceleration"

https://arxiv.org/pdf/1701.00485v2 -- "Two-Bit Networks for Deep Learning on Resource-Constrained Embedded Devices"

== VLSI ==

http://ncs.ethz.ch/pubs/pdf/Indiveri_etal06.pdf -- "A VLSI array of low-power spiking neurons and bistable synapses with spike–timing dependent plasticity"

== Pruning ==

http://papers.nips.cc/paper/5784-learning-both-weights-and-connections-for-efficient-neural-network.pdf --
"Learning both Weights and Connections for Efficient Neural Networks"

https://arxiv.org/pdf/1701.04465 -- "The Incredible Shrinking Neural Network: New Perspectives on Learning Representations Through The Lens of Pruning"

https://arxiv.org/pdf/1512.08571 -- "Structured Pruning of Deep Convolutional Neural Networks"

https://arxiv.org/pdf/1611.01427 -- "Sparsely-Connected Neural Networks: Towards Efficient VLSI Implementation of Deep Neural Networks"

== Efficient Neural Networks via Compression, Quantization, Model Reduction, etc ==

https://arxiv.org/pdf/1504.04788 -- "Compressing Neural Networks with the Hashing Trick"

https://arxiv.org/pdf/1509.08745 -- "Compression of Deep Neural Networks on the Fly"

https://arxiv.org/pdf/1502.03436 -- "An Exploration of Parameter Redundancy in Deep Networks with Circulant Projections"

https://arxiv.org/pdf/1510.00149 -- "Deep Compression: Compressing Deep Neural Networks with Pruning, Trained Quantization and Huffman Coding"

https://arxiv.org/pdf/1612.00891 -- "Parameter Compression of Recurrent Neural Networks and Degradation of Short-term Memory"

https://arxiv.org/pdf/1609.07061 -- "Quantized Neural Networks: Training Neural Networks with Low Precision Weights and Activations"

https://arxiv.org/pdf/1607.05418 -- "Runtime Configurable Deep Neural Networks for Energy-Accuracy Trade-off"

https://arxiv.org/pdf/1602.08194 -- "Scalable and Sustainable Deep Learning via Randomized Hashing"

https://arxiv.org/pdf/1508.05463 -- "StochasticNet: Forming Deep Neural Networks via Stochastic Connectivity"

https://arxiv.org/pdf/1412.7024 -- "Training Deep Neural Networks with Low Precision Multiplications"

https://arxiv.org/pdf/1612.03940 -- "Understanding the Impact of Precision Quantization on the Accuracy and Energy of Neural Networks"

https://arxiv.org/pdf/1609.00222 -- "Ternary Neural Networks for Resource-Efficient AI Applications"

== Neural Network Hyperparameter Optimization ==

https://arxiv.org/pdf/1601.00917 -- "DrMAD: Distilling Reverse-Mode Automatic Differentiation for Optimizing Hyperparameters of Deep Neural Networks"

== Neural Network based EEG Analysis ==
end
http://www.seas.upenn.edu/~wulsin/docs/wulsin2010.pdf -- "Semi-Supervised Anomaly Detection for EEG Waveforms Using Deep Belief Nets"

http://inter-eng.upm.ro/2012/files/proceedings/papers/paper72.pdf --
"Neural Network Parallelization on FPGA Platform for EEG Signal Classification"

== Seizure Detection ==

also see https://noisebridge.net/wiki/Kaggle for a (September 2016) current project!

and https://github.com/kevinjos/kaggle-aes-seizure-prediction (some earlier exploration, November 2014)

(broken link, sorry) http://www.sersc.org/journals/ijsip/vol7_no5/26.pdf --
"A Neural Network Model for Predicting Epileptic Seizures based on Fourier-Bessel Functions"

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2230664/pdf/nihms-34528.pdf --
"A Stochastic Framework for Evaluating Seizure Prediction Algorithms Using Hidden Markov Models"

(another broken link) http://cs.uni-muenster.de/Professoren/Lippe/diplomarbeiten/html/eisenbach/Untersuchte%20Artikel/PPHD+00.pdf --
"Recurrent neural network based preenddiction of epileptic seizures in intra- and extracranial EEG"

== Visible Light Sensor Network ==

http://infoteh.etf.unssa.rs.ba/zbornik/2016/radovi/KST-1/KST-1-15.pdf --
"Analysis of Visible Light Communication System for Implementation in Sensor Networks"

== Neurophysiology ==

http://www.buzsakilab.com/content/PDFs/BuzsakiKoch2012.pdf -- "The origin of extracellular fields and
currents — EEG, ECoG, LFP and spikes"

== Signal Processing ==

http://provideyourown.com/2011/analogwrite-convert-pwm-to-voltage/ -- "Arduino’s AnalogWrite – Converting PWM to a Voltage"

http://sim.okawa-denshi.jp/en/PWMtool.php -- "RC Low-pass Filter Design for PWM (Transient Analysis Calculator)"

== Hyperdimensional Computing ==

http://redwood.berkeley.edu/pkanerva/papers/kanerva09-hyperdimensional.pdf --
"Hyperdimensional Computing: An Introduction to Computing in Distributed Representation with High-Dimensional Random Vectors"

http://arxiv.org/pdf/1602.03032.pdf --
"Associative Long Short-Term Memory"

== Bird Flocks and Maximum Entropy ==

http://arxiv.org/pdf/1107.0604v1 --
"Statistical Mechanics and Flocks of Birds"

http://arxiv.org/pdf/1307.5563v1 --
"Social interactions dominate speed control in driving natural flocks toward criticality"

http://ocw.mit.edu/courses/electrical-engineering-and-computer-science/6-050j-information-and-entropy-spring-2008/syllabus/MIT6_050JS08_penfield.pdf --
"Information and Entropy (Course Notes)"

== Whale Songs ==

https://arxiv.org/pdf/1307.0589.pdf -- "The Orchive : Data mining a massive bioacoustic archive"

https://www.researchgate.net/profile/Herbert_Roitblat/publication/13429327_The_neural_network_classification_of_false_killer_whale_%28Pseudorca_crassidens%29_vocalizations/links/540d2ff60cf2df04e75478cd.pdf -- "The neural network classification of false killer whale (Pseudorca crassidens) vocalizations"

http://users.iit.demokritos.gr/~paliourg/papers/PhD.pdf -- "REFINEMENT OF TEMPORAL CONSTRAINTS IN AN EVENT RECOGNITION SYSTEM USING SMALL DATASETS"

https://www.nersc.no/sites/www.nersc.no/files/master_thesis_sebastian_menze.pdf -- "Estimating fin whale distribution from ambient noise spectra using Bayesian inversion"

http://sis.univ-tln.fr/~glotin/IJCNN2015_IHMMbioac_BartChamGlot.pdf -- "Hierarchical Dirichlet Process Hidden Markov Model for Unsupervised Bioacoustic Analysis"

https://www.inf.ed.ac.uk/publications/thesis/online/IM030057.pdf -- "Hidden Markov Model Clustering of Acoustic Data"

http://danielnouri.org/notes/2014/01/10/using-deep-learning-to-listen-for-whales/ -- Using deep learning to listen for whales

== Computational Cognitive Neuroscience ==

http://www.pnas.org/content/110/41/16390.full -- "Indirection and symbol-like processing in the prefrontal cortex and basal ganglia"

http://ruccs.rutgers.edu/images/personal-zenon-pylyshyn/docs/jaf.pdf -- "Connectionism and Cognitive Architecture: A Critical Analysis"

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2728678/pdf/nihms131814.pdf -- Neves et al 2008 "Cell Shape and Negative Links in Regulatory Motifs Together Control Spatial Information Flow in Signaling Networks"

http://psych.colorado.edu/~oreilly/papers/AisaMingusOReilly08.pdf -- "The Emergent Neural Modeling System"

https://grey.colorado.edu/emergent/index.php/Comparison_of_Neural_Network_Simulators

== Text Generation ==

http://www.cs.toronto.edu/~ilya/pubs/2011/LANG-RNN.pdf -- "Generating Text with Recurrent Neural Networks"

== Games ==

http://setgame.com/sites/default/files/teacherscorner/COGNITIVE%20MODELING%20WITH%20SET.pdf -- "How to Construct a Believable Opponent using Cognitive Modeling in the Game of Set"

http://www-personal.umich.edu/~charchan/SET.pdf -- "SETs and Anti-SETs: The Math Behind the Game of SET"

http://personal.plattsburgh.edu/quenelgt/talks/set.pdf -- "Introduction to Set"

http://web.engr.illinois.edu/~pbg/papers/set.pdf -- "On the Complexity of the Game of Set"

http://www.warwick.ac.uk/staff/D.Maclagan/papers/set.pdf -- "The Card Game Set"

http://www.math.ucdavis.edu/~anne/FQ2011/set_game.pdf -- "The Game Set"

https://www.youtube.com/watch?v=k2rgzZ2WXKo -- "Best Practices for Procedural Narrative Generation" Chris Martens

== Large Scale Brain Simulation ==

http://www.nowere.net/b/arch/96550/src/1378907656268.pdf -- "A world survey of artificial brain projects, Part I: Large-scale brain simulations"

== Music ==

http://cmr.soc.plymouth.ac.uk/publications/bci-wkshop.pdf -- "ON GENERATING EEG FOR CONTROLLING MUSICAL SYSTEMS"

== Code ==

https://github.com/nbdt/gotrain (our ANN code)

https://github.com/nbdt/openbci-golang-server (EEG data acquisition and visualization software for OpenBCI)

== Hidden Markov Models ==

http://www.cs.sjsu.edu/~stamp/RUA/HMM.pdf
-- "A Revealing Introduction to Hidden Markov Models"

http://www.jelmerborst.nl/pubs/Borst2013b.pdf
-- "Discovering Processing Stages by combining EEG with Hidden Markov Models"

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2230664/pdf/nihms-34528.pdf
-- "A Stochastic Framework for Evaluating Seizure Prediction Algorithms Using Hidden Markov Models"

http://www.schalklab.org/sites/default/files/misc/Coupled%20hidden%20markov%20model%20for%20electrocorticographic%20signal%20classification.pdf
-- "Coupled Hidden Markov Model for Electrocorticographic Signal Classification"

== Long Short Term Memory ==

http://deeplearning.cs.cmu.edu/pdfs/Hochreiter97_lstm.pdf
-- "Long Short-Term Memory"

ftp://ftp.idsia.ch/pub/juergen/2002_icannMusic.pdf
-- "Learning The Long-Term Structure of the Blues"

http://www.overcomplete.net/papers/nn2012.pdf
-- "A generalized LSTM-like training algorithm for second-order recurrent neural networks"

http://www.iro.umontreal.ca/~eckdoug/papers/2003_nn.pdf
-- "Kalman filters improve LSTM network performance in problems unsolvable by traditional recurrent nets"

http://googleresearch.blogspot.ch/2015/09/google-voice-search-faster-and-more.html
-- "Long Short-Term Memory dramatically improves Google Voice etc"

https://arxiv.org/pdf/1511.05552v4.pdf --
"Recurrent Neural Networks Hardware Implementation on FPGA"

http://vast.cs.ucla.edu/sites/default/files/publications/ASP-DAC2017-1352-11.pdf --
"FPGA-based Accelerator for Long Short-Term Memory Recurrent Neural Networks"

== Question Answering ==

http://www.overcomplete.net/papers/bica2012.pdf
-- "Neural Architectures for Learning to Answer Questions"

https://cs.umd.edu/~miyyer/pubs/2014_qb_rnn.pdf
-- "A Neural Network for Factoid Question Answering over Paragraphs"

http://arxiv.org/pdf/1502.05698.pdf
-- "Towards AI-Complete Question Answering: A Set of Prerequisite Toy Tasks"

http://www.peterschueller.com/pub/2014/2014_winograd_schemas_relevance_knowledge_graphs.pdf
-- "Tackling Winograd Schemas by Formalizing Relevance Theory in Knowledge Graphs"

http://ijcai.org/papers15/Papers/IJCAI15-190.pdf
-- "Towards Addressing the Winograd Schema Challenge — Building and Using a Semantic Parser and a Knowledge Hunting Module"

http://arxiv.org/pdf/1506.05869v2.pdf
-- "A Neural Conversational Model"

http://arxiv.org/pdf/1508.05508v1.pdf
-- "Towards Neural Network-based Reasoning"

http://www.visualqa.org/vqa_iccv2015.pdf
-- "VQA: Visual Question Answering"

== Propagators ==
Cells must support three operations:
*add some content
*collect the content currently accumulated
*register a propagator to be notified when the accumulated content changes
*When new content is added to a cell, the cell must merge the addition with the content already present. When a propagator asks for the content of a cell, the cell must deliver a complete summary of the information that has been added to it.
*The merging of content must be commutative, associative, and idempotent. The behavior of propagators must be monotonic with respect to the lattice induced by the merge operation.
*http://web.mit.edu/~axch/www/art.pdf *http://groups.csail.mit.edu/mac/users/gjs/propagators/
*http://dustycloud.org/blog/sussman-on-ai/

== Boosting ==

http://www.cs.princeton.edu/courses/archive/spr08/cos424/readings/Schapire2003.pdf
-- "The Boosting Approach to Machine Learning An Overview"

http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.19.9955&rep=rep1&type=pdf
-- "Ensembling Neural Networks: Many Could Be Better Than All"

http://www.cs.columbia.edu/~rocco/Public/mlj9.pdf
-- "Random Classification Noise Defeats All Convex Potential Boosters"

== Support Vector Machines ==

http://research.microsoft.com/en-us/um/people/cburges/papers/svmtutorial.pdf
-- "A Tutorial on Support Vector Machines for Pattern Recognition"

== Wire Length / Small World Networks ==

http://koulakovlab.cshl.edu/publications/koulakov_chklovskii_2000.pdf
-- "A wire length minimization approach to ocular dominance patterns in mammalian visual cortex"

http://papers.nips.cc/paper/1910-foundations-for-a-circuit-complexity-theory-of-sensory-processing.pdf
-- "Foundations for a Circuit Complexity Theory of Sensory Processing"

https://www.nada.kth.se/~cjo/documents/small_world.pdf
-- "Small-World Connectivity and Attractor Neural Networks"

http://www.doc.ic.ac.uk/~mpsha/ShanahanPhysRevEPreprint.pdf
-- "The Dynamical Complexity of Small-World Networks of Spiking Neurons"

http://www.dam.brown.edu/people/elie/papers/small_world.pdf
-- "Autoassociative Memory Retrieval and Spontaneous Activity Bumps in Small-World Networks of Integrate-and-Fire Neurons"

http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.332.4501&rep=rep1&type=pdf
-- "Transition from Random to Small-World Neural Networks by STDP Learning Rule"

http://tamar.tau.ac.il/~eshel/papers/J_Neural_Engineering.pdf
-- "Compact self-wiring in cultured neural networks"

== Backpropagation ==

http://page.mi.fu-berlin.de/rojas/neural/neuron.pdf -- "Neural Networks - A Systematic Introduction"

http://www4.rgu.ac.uk/files/chapter3%20-%20bp.pdf (works through simple example)

http://mattmazur.com/2015/03/17/a-step-by-step-backpropagation-example/

http://page.mi.fu-berlin.de/rojas/neural/chapter/K7.pdf

also see http://page.mi.fu-berlin.de/rojas/neural/neuron.pdf (entire book "Neural Networks - a Systemic Introduction" by Raul Rojas)

http://work.caltech.edu/lectures.html Hoeffding's inequality, VC Dimension and Back Propagation ANN

http://www.cs.stir.ac.uk/research/publications/techreps/pdf/TR148.pdf ("Learning XOR: exploring the space of a classic problem")

http://www.sysc.pdx.edu/classes/Werbos-Backpropagation%20through%20time.pdf
-- "Backpropagation Through Time: What it Does and How to Do It"

== Computer Vision ==

http://www.cv-foundation.org/openaccess/CVPR2015.py -- some interesting papers here

http://www.cv-foundation.org/openaccess/content_cvpr_2015/papers/Nguyen_Deep_Neural_Networks_2015_CVPR_paper.pdf -- "Deep Neural Networks are Easily Fooled: High Confidence Predictions for Unrecognizable Images"

== Visual Perception (Biological Systems) ==

http://cbcl.mit.edu/publications/ps/Serre_etal_PBR07.pdf
-- "A quantitative theory of immediate visual recognition"

http://www.dam.brown.edu/ptg/REPORTS/Invariance.pdf
-- "Invariance and Selectivity in the Ventral Visual Pathway"

http://www.cnbc.cmu.edu/~tai/papers/lee_mumford_josa.pdf
-- "Hierarchical Bayesian inference in the visual cortex"

== Neural Synchrony ==

http://arxiv.org/pdf/1312.6115.pdf
-- "Neuronal Synchrony in Complex-Valued Deep Networks"

== Spiking Neural Networks ==

http://ncs.ethz.ch/projects/evospike/publications/ICONIP2011%20Springer%20LNCS%20Nutta.pdf -- "EEG Classification with BSA Spike Encoding Algorithm and Evolving Probabilistic Spiking Neural Network"

http://dai.fmph.uniba.sk/~benus/confers/CNGM_ICANN05.pdf -- "Computational Neurogenetic Modeling: Integration of Spiking Neural Networks, Gene Networks, and Signal Processing Techniques"

http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.97.3902 -- "Pattern Recognition in a Bucket"

http://www.igi.tugraz.at/maass/psfiles/221.pdf -- "Noise as a Resource for Computation and Learning in Spiking Neural Networks"

http://web.stanford.edu/group/brainsinsilicon/goals.html -- Neurogrid

http://apt.cs.manchester.ac.uk/projects/SpiNNaker/ -- SpiNNaker

http://personalpages.manchester.ac.uk/staff/javier.navaridas/pubs/2012.jpdc.pdf -- more on SpiNNaker

http://apt.cs.manchester.ac.uk/ftp/pub/apt/papers/JavN_ICS09.pdf -- "Understanding the Interconnection Network of SpiNNaker"

==Hierarchical Temporal Memory==

https://github.com/numenta/nupic/wiki/Hierarchical-Temporal-Memory-Theory

http://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1000532 -- "Towards a Mathematical Theory of Cortical Micro-circuits"

==Distributed Neural Networks==

https://www.paypal-engineering.com/2015/01/12/deep-learning-on-hadoop-2-0-2/ -- paypal engineering blog, paper describes implementation of neural network as described by [http://www.maths.bris.ac.uk/~maxvd/reduce_dim.pdf Hinton 2006] on Hadoop

http://www.sciencemag.org/content/suppl/2006/08/04/313.5786.504.DC1/Hinton.SOM.pdf -- supplementary material for [http://www.maths.bris.ac.uk/~maxvd/reduce_dim.pdf Hinton 2006]

http://www.cs.otago.ac.nz/staffpriv/hzy/papers/pdcs03.pdf -- "Parallelization of a Backpropagation Neural Network on a Cluster Computer"

http://arxiv.org/pdf/1404.5997v2.pdf -- "One weird trick for parallelizing convolutional neural networks"

http://www.ics.uci.edu/~jmoorkan/pub/gpusnn-ijcnn.pdf -- "Efficient Simulation of Large-Scale Spiking Neural Networks Using CUDA Graphics Processors"

==Mixture of Experts==

http://www.cs.toronto.edu/~fritz/absps/jjnh91.pdf -- "Adaptive Mixtures of Local Experts"

==Hopfield nets and RBMs==

http://www.iro.umontreal.ca/~bengioy/papers/ftml_book.pdf -- Yoshua Bengio, 2009, Learning Deep Architectures for AI

http://deeplearning.cs.cmu.edu/ -- Syllabus for cs course on deep learning, possible source of literature for the library

https://class.coursera.org/neuralnets-2012-001/lecture -- Lecture 11 - 15 to aid in understanding of Hinton 2006 paper above

http://www.pnas.org/content/79/8/2554.full.pdf -- "Neural networks and physical systems with emergent collective computational abilities" (Hopfield 1982)

http://page.mi.fu-berlin.de/rojas/neural/chapter/K13.pdf -- "The Hopfield Model" (Rojas 1996)

http://www.seas.upenn.edu/~wulsin/docs/wulsin2010.pdf -- "Semi-Supervised Anomaly Detection for EEG Waveforms Using Deep Belief Nets"

http://www.cse.buffalo.edu/DBGROUP/bioinformatics/papers/cameraready_xwjiabibe.pdf -- "A Novel Semi-supervised Deep Learning Framework
for Affective State Recognition on EEG Signals"

http://www.cs.toronto.edu/~hinton/absps/guideTR.pdf -- "A Practical Guide to Training Restricted Boltzmann Machines"

http://arxiv.org/pdf/1503.07793v2.pdf
-- "Gibbs Sampling with Low-Power Spiking Digital Neurons"

http://arxiv.org/pdf/1311.0190v1 -- "On the typical properties of inverse problems in statistical mechanics" Iacopo Mastromatteo 2013

http://jmlr.org/proceedings/papers/v5/salakhutdinov09a/salakhutdinov09a.pdf -- "Deep Boltzmann Machines" Salakhutdinov & Hinton 2009

http://www.cs.toronto.edu/~hinton/absps/tr00-004.pdf -- "Training Products of Experts by Minimizing Contrastive Divergence"

http://www.eecg.toronto.edu/~pc/research/publications/ly.fpga2009.submitted.pdf -- "A High-Performance FPGA Architecture for Restricted
Boltzmann Machines" Ly & Chow 2009

https://pdfs.semanticscholar.org/85fa/f7c3c05388e2bcd097a416606bdd88fc0c7c.pdf -- "A MULTI-FPGA ARCHITECTURE FOR STOCHASTIC RESTRICTED BOLTZMANN MACHINES" Ly & Chow 2009

== Variational Renormalization ==

https://arxiv.org/pdf/1410.3831 -- "An exact mapping between the Variational Renormalization Group and Deep Learning"

== Neuromorphic Stuff ==

https://arxiv.org/pdf/1508.01008.pdf -- "INsight: A Neuromorphic Computing System for Evaluation of Large Neural Networks" Chung, Shin & Kang 2015

== Markov Chain Monte Carlo ==

http://www.cs.princeton.edu/courses/archive/spr06/cos598C/papers/AndrieuFreitasDoucetJordan2003.pdf
-- "An Introduction to MCMC for Machine Learning"

http://jmlr.org/proceedings/papers/v37/salimans15.pdf
-- "Markov Chain Monte Carlo and Variational Inference: Bridging the Gap"

https://www.umiacs.umd.edu/~resnik/pubs/LAMP-TR-153.pdf -- "Gibbs Sampling for the Uninitiated"

==Entrainment==

http://www.scirp.org/Journal/PaperDownload.aspx?paperID=33251 -- "Alpha Rhythms Response to 10 Hz Flicker Is Wavelength Dependent"

http://www.brainmachine.co.uk/wp-content/uploads/Herrmann_Flicker.pdf -- "EEG responses to 1–100 Hz flicker: resonance phenomena in visual cortex and their potential correlation to cognitive phenomena"

http://www.jneurosci.org/content/23/37/11621.full.pdf -- "Human Cerebral Activation during Steady-State Visual-Evoked Responses"

http://www.dauwels.com/Papers/CogDyn%202009.pdf -- "On the synchrony of steady state visual evoked potentials and oscillatory burst events"

https://www.tu-ilmenau.de/fileadmin/public/lorentz-force/publications/peer/2012/haueisen2012/Halbleib_JCN_2012_Topographic_analysis_photic_driving.pdf -- "Topographic Analysis of Engagement and Disengagement of Neural Oscillators in Photic Driving: A Combined Electroencephalogram/Magnetoencephalogram Study"

==Mining Scientific Literature==

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC153503/pdf/1471-2105-4-11.pdf -- "PreBIND and Textomy – mining the biomedical literature for protein-protein interactions using a support vector machine" Donaldson 2003 BMC Bioinformatics

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4674139/pdf/pcbi.1004630.pdf -- "Text Mining for Protein Docking" Badal 2015 PLoS Comput Biol.

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4691339/pdf/bav116.pdf -- "Biocuration with insufficient resources and fixed timelines" Rodriguez-Esteban 2015 Database: The Journal of Biological Databases and Curation

==(not necessarilly very) Current Discussion==

re Tononi's "Integrated Information Theory" http://www.scottaaronson.com/blog/?p=1799

(19 February 2014) starting to think about possibility for experiments (loosely) related to [https://en.wikipedia.org/wiki/Visual_evoked_potential Visual Evoked Potential] research again - for instance:

http://www.biosemi.com/publications/pdf/Bierman.pdf -- Wave function collapse

[[File:Schak99InstantaneousCoherenceStroopTask.pdf]] -- "Instantaneous EEG Coherence Analysis During the Stroop Task" -- Schack et al 1999

[[File:CoherentEEGAmbiguousFigureBinding.pdf]] -- "Coherent EEG Indicators of Cognitive Binding During Ambiguous Figure Tasks" -- Klemm, Li, and Hernandez 2000

Note these two papers flog coherence measures - not trying to focus so much on that analysis right now, more interested in general understanding of what these experiments are about with possible goal of designing simpler experiments & analysis of similar perceptual/cognitive phenomena.

Here is an article that looks more directly at visual evoked potential measures:

[[File:ERP_Stereoscopic.pdf]] -- "Evoked Responses to Distinct and Nebulous Stereoscopic Stimuli" -- Dunlop et al 1983

(11 September 2013) more on analysis methods:

http://slesinsky.org/brian/misc/eulers_identity.html

http://www.dspguide.com/ch8/1.htm

[[File:Fftw3.pdf]]

[[File:ParametricEEGAnalysis.pdf]]

[[File:ICATutorial.pdf]]

[[File:ICAFrequencyDomainEEG.pdf]]

(21 August 2013) - readings relating statistical (etc math / signal processing / pattern recognition / machine learning) methods for EEG data interpretation. A lot of stuff, a bit of nonsense ... and ... statistics!

Would be good to identify any papers suitable for more in-depth study. Currently have a wide field to graze for selections:

[[File:DWTandFFTforEEG.pdf]] "EEG Classifier using Fourier Transform and Wavelet Transform" -- Maan Shaker, 2007

[[File:Schak99InstantaneousCoherenceStroopTask.pdf]] -- "Instantaneous EEG Coherence Analysis During the Stroop Task" -- Schack et al 1999

[[File:KulaichevCoherence.pdf]] -- "The Informativeness of Coherence Analysis in EEG Studies" -- A. P. Kulaichev 2009 ''note: interesting critical perspective re limitations, discussion of alternative analytics''

[[File:ContinuousAndDiscreteWaveletTransforms.pdf]] -- review of (pre-1990) wavelet literature -- Christopher Heil and David Walnut, 1989

[[File:EEGGammaMeditation.pdf]] -- "Brain sources of EEG gamma frequency during volitionally meditation-induced, altered states of consciousness, and experience of the self" -- Dietrich Lehman et al 2001

http://neuro.hut.fi/~pavan/home/Hyvarinen2010_FourierICA_Neuroimage.pdf - "Independent component analysis of short-time Fourier transforms for spontaneous EEG/MEG analysis" -- Aapo Hyvarinen, Pavan Ramkumar, Lauri Parkkonen, Riitta Hari - paper published in Neuroimage vol 49 (2010)

----

[http://www.nickgillian.com/software/grt OpenSource Machine Learning Algs from NG @MIT]
 [https://www.usenix.org/system/files/conference/usenixsecurity12/sec12-final56.pdf Consumer grade EEG used to see "P300" reponse] and for thoes with a short attention span [http://www.extremetech.com/extreme/134682-hackers-backdoor-the-human-brain-successfully-extract-sensitive-data tldr]
 (discussed at meetup Wednesday 31 July 2013)
 "Coherent EEG Indicators of Cognitive Binding During Ambiguous Figure Tasks" Klemm, Li, and Hernandez 2000
 [[File:CoherentEEGAmbiguousFigureBinding.pdf]]
 "We tested the hypothesis that perception of an alternative image in ambiguous figures would be manifest as high-frequency (gamma) components that become synchronized over multiple scalp sites as a "cognitive binding" process occurs."
 

----

[http://dreamsessions.net art, dream, and eeg]

----

[http://www.believermag.com/issues/200710/?read=article_aviv mind v brain, hobson v solms]
 http://en.wikipedia.org/wiki/Activation-synthesis_hypothesis
 [[File:HobsomREMDreamProtoconsciousness.pdf|Hobson09ProtosconsciousnessREMDream]]

"Hobson and McCarley originally proposed in the 1970s that the differences in the waking-NREM-REM sleep cycle was the result of interactions between aminergic REM-off cells and cholinergic REM-on cells.[4] This was perceived as the activation-synthesis model, stating that brain activation during REM sleep results in synthesis of dream creation.[1][1] Hobson's five cardinal characteristics include: intense emotions, illogical content, apparent sensory impressions, uncritical acceptance of dream events, and difficulty in being remembered."

----

Berkeley Labs

[http://gallantlab.org/index.html Gallant Group]
 [http://walkerlab.berkeley.edu/ Walker Group]
 [http://socrates.berkeley.edu/~plab/ Palmer Group]

==Sleep Research==

[http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2335403/ Comment on the AASM Manual for the Scoring of Sleep and Associated Events]

==random tangents==
(following previous discussion) - we might select a few to study in more depth
(... or not! Plenty more to explore - suggestions (random or otherwise) are welcome.
http://www.meltingasphalt.com/neurons-gone-wild/ --
Neurons Gone Wild - Levels of agency in the brain.

'''stereoscopic perception:'''
*[[File:ERP_Stereoscopic.pdf]]

some (maybe) interesting background on Information Theory (cool title...)
Claude Shannon: "Communication in the Presence of Noise"
[[File:Shannon_noise.pdf]]
"We will call a system that transmits without errors at the rate ''C'' an ideal system.
Such a system cannot be achieved with any finite encoding process
but can be approximated as closely as desired."

wikipedia etc quick reads:
https://en.wikipedia.org/wiki/Eeg
https://en.wikipedia.org/wiki/Neural_synchronization
https://en.wikipedia.org/wiki/Event-related_potentials
http://www.scholarpedia.org/article/Spike-and-wave_oscillations
http://www.scholarpedia.org/article/Thalamocortical_oscillations

==Previously==

[http://www.psychiclab.net/ Masahiro's EEG Device/IBVA Software]

[http://www.instructables.com/id/open-brain-wave-interface-hardware-1/ and ... open source hardware design and kits on instructables.com]

[http://brainstorms.puzzlebox.info/ Puzzlebox - Opensource BCI Developers]

Morgan from GazzLab @ MissionBay/UCSF

https://github.com/gazzlab

Let's ease into a lightweight "journal club" discussion with this technical report from NeuroSky.

Name: A user-friendly SSVEP-based brain-computer interface using a time-domain classifier, Luo A and Sullivan TJ 2010

URL: [[File:NeuroSkyVEP.pdf‎]]

Please add your comments & questions here.

==Background Reading==

http://nanosouffle.net/ (view into Arxiv.org)

Name: Hunting for Meaning after Midnight, Miller 2007

URL: <http://www.noisebridge.net/pipermail/neuro/attachments/20130501/4b992eb6/attachment-0002.pdf>

Name: Broken mirrors, Ram, VS, & Oberman, LM, 2006, Nov

URL: <http://www.noisebridge.net/pipermail/neuro/attachments/20130501/4b992eb6/attachment-0003.pdf>

Ramachandran Critique

http://blogs.scientificamerican.com/guest-blog/2012/11/06/whats-so-special-about-mirror-neurons/

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2773693/

Sleep/Dream Studies

http://www.cns.atr.jp/dni/en/publications/

==NeuroSky Docs==
[[File:NeuroSkyDongleProtocol.pdf‎]]

[[File:NeuroSkyCommunicationsProtocol.pdf‎]]

==Android Neutral Network Fuzzy Learning app==
[https://play.google.com/store/apps/details?id=com.faadooengineers.free_neuralnetworkandfuzzysystems Android Neutral Network Fuzzy Learning app in Play Store]

==Learning about Neural Networks==
* What type of network? [http://stats.stackexchange.com/questions/114385/what-is-the-difference-between-convolutional-neural-networks-restricted-boltzma RMB (Restricted Boltzmann Machine) vs Autoencoder/MLP vs CNN (Convolutional Neural Networks)]
* Andrej Karpathy's [http://cs.stanford.edu/people/karpathy/convnetjs/ Convolutional Neural Network coded in JavaScript (ConvNetJS)]
* Andrej Karpathy's [http://karpathy.github.io/2015/10/25/selfie/ What a Deep Neural Network thinks about your #selfie (background on Convolutional Neural Networks for image recognition and classification)]
* [https://blog.webkid.io/neural-networks-in-javascript/ Neural Networks in JavaScript w/MNIST]
* [http://www.antoniodeluca.info/blog/10-08-2016/neural-networks-in-javascript.html Another NN in JS]
* [http://caza.la/synaptic/ The Synaptic "architecture-free" neural network library in JS]

DreamTeam

2018-04-13T05:39:29Z

Bambikill: /* Literature & Other Resources */

== Meetings ==
Wednesday 8pm

== Hardware ==
NEW Donation:

<big>'''DreamTeam just received a new Emotiv head scanning doobie hobber!'''</big>

[[DreamTeam/Valis]]

[[DreamTeam/Brainduino]]

[[DreamTeam/flashingOpenBCIRadios]]

[http://op-innovations.com/ TrueSense]
 [http://www.openbci.com OpenBCI]
 [http://psychiclab.net/IBVA/PsychicLab.html Psychic Labs IBVA]
 [http://www.instructables.com/id/open-brain-wave-interface-hardware-1/ Masahiro Instructables]
 Neurosky Mindwave
 Emotiv Epoc
 [http://www.Open-BCI.org wWw.Open-BCI.org]
 https://noisebridge.net/wiki/OpenEEG

== Code ==

Some set game code hosted here: https://wiki.journalclub.info/

[[DreamTeam/pastebin]]

[[File:Plex.txt]] data acquisition for neurosky mindwave - starting to shape up ...

[[Polyhedrons and Perceptrons]]

== Upcoming Field Trip ==

(rescheduled ... will try to make it out next week - September 2nd)
http://www.meetup.com/NeuroTechSF/ on August 26, 2015

== Literature & Other Resources ==
''' see [[DreamTeam/Reading]] for a comprehensive list of topics and articles of interest'''

[https://sites.google.com/site/wmvanvliet/tutorials/eeg-signal-analysis EEG Data Analysis tutorial with Marijn van Vliet] and [https://github.com/wmvanvliet/neuroscience_tutorials MvV github tutorials]

[http://www.cns.atr.jp/dni/en/downloads/neuralcodeconverter-3/ Neural Code Converter (see below May 2013 for previous)]

[http://www.dreambank.net/ dreambank - UCSC]

[http://www.opensourceinstruments.com/Electronics/A3019/EEG.html Electronics of EEG -- mind bending thought experiments]

[https://noisebridge.net/images/9/91/NonlinearDynamicsEEGPerspectives.pdf Nonlinear Dynamics in EEG Analysis: Disappointments and Perspectives]

CHAOS! -> http://collections.infocollections.org/ukedu/uk/d/Jh0699e/4.html
* Is the EEG a strange attractor? Brain stem neuronal discharge patterns and electroencephalographic rhythms

http://sulcus.berkeley.edu/freemanwww/manuscripts/id1/88.html
* Strange Attractors that Govern Mammalian Brain Dynamics Shown by Trajectories of Electroencephalographic (EEG) Potential

^ nice graphics ... more on chaos-based models at http://sulcus.berkeley.edu ->
* ("Neuroscience" link on sidebar yields many publications from that lab]
<pre>
We use nonlinear mapping and multidimensional scaling into 2-space to identify itinerant chaotic trajectories
through sequences of nonconvergent attractor ruins in the attractor landscapes of brain state space.
The attractors are created and modified by reinforcement learning based on classical and operant conditioning.
Neural modeling includes demonstrations of the neural mechanisms that ensure in all cortices the stability of
neural populations that supports rapid and widespread state transitions by virtue of small-world and scale-free
network architecture. Neocortex is unique among cortices in maintaining global self-organized criticality,
in which the critical order parameter is the global level of neural synaptic interaction that everywhere locally
is homeostatically regulated by neural thresholds and refractory periods.
</pre>

cellular automata and non-linear dynamics -> https://theory.org/complexity/cdpt/html/node4.html

-----
chirp!
[[Image:Chirp_animation.gif]]
-----

 Currently [https://github.com/kevinjos/openbci-golang-server hacking] the [http://openbci.com OpenBCI EEG] powered by TI's [http://www.ti.com/lit/ds/symlink/ads1299.pdf ADS1299].
 [https://www.youtube.com/watch?v=HEbyITI13zg G. Tononi 'In Search of Sleep Function']
 [http://journal.frontiersin.org/Journal/10.3389/fninf.2013.00036/full interactive data visualization in Python]
 [http://lucidity.com/ Lucidity Institute]
 [http://www.dtic.upf.edu/~ralph/ClinNeurophysiol116569.pdf On the predictability of epileptic seizures]
 [https://cours.etsmtl.ca/sys828/REFS/A1/Fawcett_PRL2006.pdf ROC curve review]
 [http://www.dspguide.com/ Free book on digital signal processing]
 [http://materials.jeremybejarano.com/MPIwithPython MPIwithPython solid intro]
 [http://www.hindawi.com/journals/cin/2011/406391/ pyeeg whitepaper]
 [http://www.bearcave.com/misl/misl_tech/wavelets/hurst/ reading on the Hurst exponent, nice pictures]
 [http://www.mc.vanderbilt.edu/documents/neurology/files/Lecture%205-%20EEG%20in%20focal%20epilepsies-handout.pdf EEG in focal epilepsies]
 We've been reading some papers the past two weeks on the [http://researcher.watson.ibm.com/researcher/view_person_subpage.php?id=4700 TrueNorth processor].
 If you're interested in getting involved with the IBM led research group, [http://www.research.ibm.com/cognitive-computing/chip_form.shtml apply here].

 Electric Fields of the Brain: The Neurophysics of EEG, 2nd Edition, Paul L. Nunez & Ramesh Srinivasan
 Rhythms of the Brain, Gyorgy Buzsaki
 [http://www.cogsci.ucsd.edu/research/profile/447/ Bradley Voytek UCSD]
 [http://www.nature.com/neuro/journal/vaop/ncurrent/full/nn.3719.html Voss et al., Induction of self awareness in dreams through frontal low current stimulation of gamma activity]https://www.youtube.com/watch?v=HEbyITI13zg
 [http://www.blogs.uni-mainz.de/fb05philosophie/files/2013/04/Voss-et-al-2013_Measuring-Consciousness-in-Dreams.pdf Voss et al., Measuring consciousness in dreams: The lucidity and consciousness in dreams scale]

[[mHealth_Hackathon]]
 [https://orbit.puzzlebox.info/tracker/wiki Puzzlebox repo/tracker]
 [http://brainliner.jp EEG Data Repository]

GDF Standards:
 [http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3061298/ GDF/Biosig paper]
 [http://arxiv.org/abs/cs.DB/0608052 GDF Data Standards]
 [http://www.bcistandards.org/ Tobi Project => BCI Standards]

European Data Format:
 http://www.edfplus.info/downloads/index.html
 https://github.com/breuderink/eegtools
 http://www.neurotraces.com/edf/

== Other Groups ==

[https://class.coursera.org/neuraldata-001/ Free course on Exploring Neural Data]
[http://diy-neuro.com/ Cognitive Technology Group -- UC-Berkeley]

== IBVA Port ==

Sucessfully built and ran IVBA source (Aug 11, 2013). Key steps below.

1. Root directory of application must be /Developer/Examples/IBVAV4F5Apr12all
 2. [http://cocoabob.net/?p=1032 Installing Xcode 3.2.6 on Mountian Lion] and don't try taking shortcuts... 3.2.1 crashes on run-time when trying to load the project.

== Data ==

We have 650MB of data taken from [[party]] (Aug. 10, 2013).. Where to upload?
 [[File:20130516233839.txt]]
 [[File:20130516233839-1.txt]]

== Events on the Horizon ==

Let's have an excellent installation to bring to [http://toorcamp.toorcon.net/ Toor camp]. Recursive reentry?

== Events beyond the Horizon ==

[http://neurogaming.brainihack.org/projects/project-list/ NeuroGaming Hackathon, May, San Francisco]

[https://sudoroom.org/ai1ec_event/hackyourphd-open-knowledge-meetup/?instance_id=66319, HackYourPhd, SudoRoom, Thursday 7/25]

== (Old) Meeting Notes ==

(still meeting regularly, just lax about taking/posting notes lately. Here's some minutes from our first few meetings ...)

=== March 17, 2017 ===

Discussed findings from XTech Conference [https://www.xtechexpo.com] which led to a demo of hypnosis.

=== July 15, 2015 ===

In addition to hooking someone up to the OpenBCI and measuring voltages from grinding the jaw, a new Facebook page for the group was created: [https://www.facebook.com/groups/1626036630970351/]

=== March 26, 2014 ===

Notes from March, 26 Meeting. Discussion of dark mind hacks - to be identified and squelched. Infiltration, dissruption, the agent provocatour, etc etc...
 http://darkpatterns.org/

=== June 19, 2013 ===

[http://www.meetup.com/braindjvj/events/119780592/ Meetup@TechShop, 2013/06/19]

=== May 15, 2013 ===

"journal club" discussion to better understand a technical report from NeuroSky - see [[DreamTeam/Reading#Current_Discussion]]

=== May 8, 2013 ===
a small group studied data from the MindWave device and [https://github.com/nbdt hacked on code] to handle (so-called) "timestamps" more realistically ...

=== May 1, 2013 ===
Present: Johny, Daglan, John, Thomas, Aly, Jade, Dan, Kevin, Rohan

Recorded: https://www.noisebridge.net/wiki/File:SampleEEGdata20130501.txt

Understanding dreams via EEG readout
Nightmare or not classifier, a good place to start

delta 1-4Hz
theta 4-7Hz
alpha 6-12Hz
low beta 12-15Hz normal focus
mid beta 16-22Hz high focus
high beta 22-24Hz anxiety muscle artifact shortcoming of eeg

peak freq marker of stress

Rohan will look up information on EEG

There will be a burning man device that will be controlled by an EEG headset

[http://www.brainocate.com/ Brain-o-cate] --> interactive video game focused on teaching children about the brain

[http://www.myzeo.com/sleep/ Zeo] headband EEG made for sleep

We like more sensors and high data quality and comfort

Computational neuro science on coursera

Signal localization
*Grid array sensor
*Seizure localization 164/256 sensors on the brain, 1/2" apart
*32 channel EEG on the surface to get a picture
**ElectroCap has cap with 64 sensor places

EEG correlate for REM
*States
http://brainstorms.puzzlebox.info/
LEDs on the HSNPs
*Glow different colors by brain state

Convert brain waves to audio spectrum
*MIDI tones, 4-8Hz offset ==> induce activity
*Musical Insturment Digital Interface
**Feeback loops
**Would have to compensate for delay between the reading device and audio device

Classify audio waves via brain monitoring

What is the difference between vision/hearing and dreaming or thinking on music in the brain?
*dtMRI - detects neuronal activity, by neuron strands

Thermal infrared camera to monitor blood-flow in the brain

IR cameras to measure full specture, rather than stricly amplitude
Could construct a small array of infrared sensors -- more economic

==Project Ideas==

Send EEG data recorded during the meeting
*Guess who was wearing the device
*See Attached

Rohan has 9 months of EEG data from Hymalayas
*Has built a meditation classifier from raw Neuro Sky data

Dan would like to have real-time display, in form of LEDs on cap, that reflects EEG data
*Feedback loop between two people wearing LED caps

Brian with badasslighting and [http://www.americansteelstudios.com/2013/department-of-spontaneous-combustion-awarded-burning-man-grant dept of spontaneous combustion]
host open night in W Oakland, teach welding, art, construction
steel walk-in structure with neurons and synapses of LEDs in acrlyic tubes
0.3W LEDs all over the place
raspberry pi
flame effects for quiet mind
"processing" python tool to mix RP with picture
looking for a developers to workout the details of nero-data to brain lights
American Steel, Mandela and 20th, open house
Contact Brian if interested @ 415-509-2100/bkrawitz@gmail.com

Johny wants to convert brain data into music
*translation technique
*dream sounds outside the HSNPs

Space craft does ambient electronic music

Max Queue is an astronaut headed band

==Background==

https://www.noisebridge.net/pipermail/neuro/2013-April/000791.html

[Neuro] Coming soon to Noisebridge - The Dream Team
bfb at riseup.net
Wed Apr 24 18:00:00 UTC 2013

Previous message: [Neuro] Neuro Digest, Vol 31, Issue 3
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

Hello Noisebridge!

We shall meet on Wednesday, May 1, at 6:30, because we're all stars!
Python class is over, now let's put that good know-how to work on some
maths and science stuffs. There may also be comestibles, just this once.

Come down to the 6:30 share-a-thon. Discuss your dreams and other
applications of neuron-monitoring devices. I want to investigate lexical
dream content and brainwaves of people all over the planet, starting
with Noisebridge. Thus far, I have an EEG device and a semi-incomplete
research facility (HSNPs).

To get up to speed with what I've been looking into read on. And all you
varied neuro-hackers out there please do go to the brainliner.jp site,
it's amazing.

I've attached a copy of a sample data set taken while writing this
message. Also attached is a screen shot of the GUI that the python
scripts create. The scripts run great on Debian. The dongle that comes
with the device has an antenna to pick-up the WiFi transmissions from
the device. It opens the right sockets, makes a few hand shakes, and
promptly peaces out. Neurosky purports the device will give a good
signal up to 10m. It spews ~500 measures per second.

https://github.com/akloster/python-mindwave

This is the EEG device. Specs included.

http://store.neurosky.com/products/mindwave-1

And here's a link to Yukiyasu Kamitani's landing page with ATR, Advanced
Telecommunications Research Institute International. He's head of the
Department of Neuroinfomatics, and lead reacher on the publication
below. They are working on predicting dream sequences from EEG and MRI
data taken while asleep.

http://www.cns.atr.jp/dni/en/members/kamitani_e/
http://www.cns.atr.jp/dni/en/publications/

Kanitani and others are open-sourcing a good bit of their work. See git
hub repository and standardized repository for sharing neuro-data below.
They're set-up to run in Mat Lab, but w/e.

Really, go to brainliner and then giggle crescendoing into mad-scientist
roar.

http://brainliner.jp/
https://github.com/ATR-DNI/VBCCA

And finally, to pass on a message from a friend, because it's open in my
browser and seems appropriate.
https://raw.github.com/videlalvaro/gifsockets/master/doc/mybrain.gif

See you Wednesday.

-Kevin

-----

[[Category:Neuro]]

File:Brain-duino circuit.png

2017-03-22T03:08:44Z

Bambikill:

File:IMG 9573.jpg

2016-12-22T04:06:36Z

Bambikill: IC7?

IC7?

File:IMG 9574.jpg

2016-12-22T04:05:35Z

Bambikill: Brainduino 12/21/2016

Brainduino 12/21/2016

Talk:DreamTeam

2016-12-22T03:56:39Z

Bambikill: Upload images of brainduino

==Brainbuino images==
[[File:IMG_9574.jpg]]
[[File:IMG_9573.jpg]]

==Discussion 14 December 2016==
Restricted Boltzmann Machine
* is RBM not an RNN ?
* summary recollection - RBM is a type of NN.
Has layers. Input, output, hidden. Can have multiple hidden layers.
Has nodes and weights.

What is a "Recurrent" Neural Network?
Why not "Recursive"?

How is Error Backpropogation different concept from "Recurrent"?

Network architecture ...
adjust weights within rather than redefine network topology

RNN may have generative capacity.

Recollections of current comprehension:
* it is restricted. ie connections not allowed within layer

RBM involves MCMC (Markov chain Monte Carlo) process
via Gibbs Sampling

relates to Ising Model
Hopfield nets

==Discussion 07 October 2015==
Distributed neural networks
* hadoop - useful to handle reliability when dealing with thousands of servers
* YARN yet another resource negotiator
* apache - also mahout
* task oriented - wants to solve actual data analysis quickly - not energy efficient
* compare to BOINC or folding@home etc
* folding@home 40 petaflops - how might this support simulation of something like 10**11 neurons
* compare to spiNNaker - specialized network for thousands of ARM cores, designed for spiking neural networks but can support traditional backpropagation

==Hacking Hardware==
Presentation @ Black Hat 2011, discusses JTAG
https://media.blackhat.com/bh-dc-11/Grand/BlackHat_DC_2011_Grand-Workshop.pdf

==Restricted Boltzmann Machine==
* roots in Hopfield network
* relevance of pruning - restricted connectivity makes training feasible for non-trivial number of neurons

-----

https://github.com/search?q=neurosky&type=&ref=simplesearch (found 48 repository results)

https://github.com/nbdt == NB dream team

<pre>
// Audio Spectrum Display
// Copyright 2013 Tony DiCola (tony@tonydicola.com)

// This code is part of the guide at http://learn.adafruit.com/fft-fun-with-fourier-transforms/

#define ARM_MATH_CM4
#include <arm_math.h>
#include "FastLED.h"

// ebw code
#include <avr/pgmspace.h> // PROGMEM
#include <Cycler.h>
#include <EightBitWaves.h>

//#define LED_COUNT 13
//#define LED_CLASS WS2812
//#define LED_COLOR_ORDER GRB
#define LED_MAX_BRIGHTNESS 64 // 1/8
//#define LED_DT 2 // SERIAL DATA PIN
//#define LED_CK 6 // SERIAL CLOCK PIN
//#define SERIAL_BAUDRATE 9600
#define SERIAL_TIMEOUT 5

//Ticker ticker;
//Cycler hue_cycler, brightness_cycler;
Cycler brightness_cycler;
//struct CRGB pixel, fastled_buffer[LED_COUNT];

uint8_t minWave = 0, maxWave = 255;
uint32_t ticks_per_second = 1000;
uint32_t prevMillis = millis();
// ebw code

////////////////////////////////////////////////////////////////////////////////
// CONIFIGURATION
// These values can be changed to alter the behavior of the spectrum display.
////////////////////////////////////////////////////////////////////////////////

#define DEBUG false
#define OUTPUT_LED_DATA false
#define FULL_SET false

#define BAUD_RATE 38400

#define NUM_LEDS 12
#define DATA_PIN 2

#define ARRAY_SZ(x) (sizeof(x) / sizeof((x)[0]))

uint SAMPLE_RATE_HZ = 9000; // Sample rate of the audio in hertz.
float SPECTRUM_MIN_DB = 30.0; // Audio intensity (in decibels) that maps to low LED brightness.
float SPECTRUM_MAX_DB = 60.0; // Audio intensity (in decibels) that maps to high LED brightness.

const int FFT_SIZE = 256; // Size of the FFT. Realistically can only be at most 256
// without running out of memory for buffers and other state.
const int AUDIO_INPUT_PIN = 14; // Input ADC pin for audio data.
const int ANALOG_READ_RESOLUTION = 10; // Bits of resolution for the ADC.
const int ANALOG_READ_AVERAGING = 16; // Number of samples to average with each ADC reading.

const int ONBOARD_LED_PIN = 13; // Output pin for power LED (pin 13 to use Teensy 3.0's onboard LED).

const int MAX_CHARS = 65; // Max size of the input command buffer

uint8_t brightness = 128;

CRGB leds[NUM_LEDS];

////////////////////////////////////////////////////////////////////////////////
// INTERNAL STATE
// These shouldn't be modified unless you know what you're doing.
////////////////////////////////////////////////////////////////////////////////

IntervalTimer samplingTimer;
float samples[FFT_SIZE*2];
float magnitudes[FFT_SIZE];
int sampleCounter = 0;
char commandBuffer[MAX_CHARS];
float frequencyWindow[NUM_LEDS+1];

boolean ledState = false;

uint loopCounter;

////////////////////////////////////////////////////////////////////////////////
// setup
////////////////////////////////////////////////////////////////////////////////

void setup() {
// Set up serial port.
Serial.begin(BAUD_RATE);
delay(1000);

loopCounter = 0;

// Set up ADC and audio input.
pinMode(AUDIO_INPUT_PIN, INPUT);
analogReadResolution(ANALOG_READ_RESOLUTION);
analogReadAveraging(ANALOG_READ_AVERAGING);

// Turn on the power indicator LED.
pinMode(ONBOARD_LED_PIN, OUTPUT);
digitalWrite(ONBOARD_LED_PIN, LOW);

FastLED.addLeds<NEOPIXEL, DATA_PIN>(leds, NUM_LEDS);

// Clear the input command buffer
memset(commandBuffer, 0, sizeof(commandBuffer));

// Initialize spectrum display
spectrumSetup();
// Serial.println("** setup() spectrumSetup **********************");
// delay(1000);

brightness_cycler.setup((float) 6.0, ticks_per_second);

// Begin sampling audio
samplingBegin();
}

////////////////////////////////////////////////////////////////////////////////
// loop
////////////////////////////////////////////////////////////////////////////////

void loop() {
// Calculate FFT if a full sample is available.
if (samplingIsDone()) {
processFFT();
}
updateBrightness();
}

////////////////////////////////////////////////////////////////////////////////
// UTILITY FUNCTIONS
////////////////////////////////////////////////////////////////////////////////
void updateBrightness() {
uint16_t delta_ticks = (uint16_t)(millis() - prevMillis);
prevMillis += delta_ticks;
brightness_cycler.update(delta_ticks);
// unsigned char brightness = EightBitWaves::sine(brightness_cycler.phase(), minWave, maxWave);
brightness = EightBitWaves::sine(brightness_cycler.phase(), minWave, maxWave);
}

void processFFT() {

// Run FFT on sample data.
arm_cfft_radix4_instance_f32 fft_inst;
// These functions have been depricated
arm_cfft_radix4_init_f32(&fft_inst, FFT_SIZE, 0, 1);
arm_cfft_radix4_f32(&fft_inst, samples);

// arm_cfft_f32(&fft_inst, samples, 0, 1);

// Calculate magnitude of complex numbers output by the FFT.
arm_cmplx_mag_f32(samples, magnitudes, FFT_SIZE);

updateLEDs();

// Restart audio sampling.
samplingBegin();
}

void outputFFTData() {

if (DEBUG && FULL_SET) {
Serial.println("** FFT Data ****************************");
Serial.print("** Total size of magnitude array: ");
Serial.println(sizeof(magnitudes));

Serial.print("** Number of elements in magnitude: ");
Serial.println(sizeof(magnitudes)/sizeof(*magnitudes));

// Serial.print("** Number of elements in magnitude: ");
// Serial.println(sizeOfFloatArray(magnitudes));
}

// for (int i = 0; i < sizeof(magnitudes)/sizeof(*magnitudes); ++i) { // was FFT_SIZE
for (size_t i = 0; i < ARRAY_SZ(magnitudes); ++i) {
float intensity = magnitudes[i];
intensity = 20.0*log10(intensity);

// Scale the intensity and clamp between 0 and 1.0.
intensity -= SPECTRUM_MIN_DB;
intensity = intensity < 0.0 ? 0.0 : intensity;
intensity /= (SPECTRUM_MAX_DB-SPECTRUM_MIN_DB);
intensity = intensity > 1.0 ? 1.0 : intensity;

if (FULL_SET) {
if (DEBUG) {
Serial.print("FFT[");
Serial.print(i);
Serial.print("] = ");
Serial.println((byte)(254 * intensity));
} else {
Serial.write((byte)(254 * intensity));
}
}
}

if (FULL_SET) {
if (DEBUG) {
Serial.println("** End FFT Results **********************");
} else {
Serial.write(255);
}
}
}

/*
if (DEBUG) {
Serial.print("FFT[");
Serial.print(i);
Serial.print("] = ");
Serial.println(magnitudes[i]);
} else {
// Serial.write(); // send out the data
Serial.write(255); // Send stop Byte
}
}
*/

//**************************************************************************
// updateGrnLED
//**************************************************************************
//
void updateGrnLED() {
// The measured time between two consecutive events (rock solid on the scope):
// FHT_N mSec Baud Rate
//
// Serial.print("updateOnboardLED");

digitalWrite(ONBOARD_LED_PIN, HIGH && ledState); // turn the LED on or off (HIGH is the voltage level)
ledState = !ledState;
}

//**************************************************************************
// frequencyToBin
// Function - Convert a frequency to the appropriate FFT bin it will fall within.
//**************************************************************************
//
int frequencyToBin(float frequency) {
float binFrequency = float(SAMPLE_RATE_HZ) / float(FFT_SIZE);
return int(frequency / binFrequency);
}

//**************************************************************************
// updateLEDs
// Function - Processes the FFT data and outputs to LEDs. Update each LED based
// on the intensity of the audio in the associated frequency window.
//**************************************************************************
//
void updateLEDs() {
float intensity, otherMean;

if (OUTPUT_LED_DATA && DEBUG)
Serial.println("** Start LED FFT Results **********************");

// Serial.print("ARRAY_SZ(leds) = ");
// Serial.print(ARRAY_SZ(leds));
// Serial.print(", sizeof magnitudes = ");
// Serial.println(ARRAY_SZ(magnitudes));

for (size_t i = 0; i < ARRAY_SZ(leds); ++i) {
// windowMean(magnitudes,
// frequencyToBin(frequencyWindow[i]),
windowMean(frequencyToBin(frequencyWindow[i]),
frequencyToBin(frequencyWindow[i+1]),
&intensity,
&otherMean);

// Serial.print("**1 intensity[");
// Serial.print(i);
// Serial.print("] = ");
// Serial.println(intensity);

// Convert intensity to decibels.
intensity = 20.0*log10(intensity);

// Scale the intensity and clamp between 0 and 1.0.
intensity -= SPECTRUM_MIN_DB;
intensity = intensity < 0.0 ? 0.0 : intensity;
intensity /= (SPECTRUM_MAX_DB-SPECTRUM_MIN_DB);
intensity = intensity > 1.0 ? 1.0 : intensity;

// Serial.print("**2 intensity[");
// Serial.print(i);
// Serial.print("] = ");
// Serial.println(intensity);

// Output intensity value to LED
float newHue = 255*intensity;
if (newHue < 64) {
newHue = 0;
} else {
newHue = map(newHue, 64, 255, 0, 255);
}

leds[i] = CHSV(newHue, 255, brightness);

if (OUTPUT_LED_DATA) {
if (DEBUG) {
Serial.print("FFT[");
Serial.print(i);
Serial.print("] = ");
Serial.println(255.0 * intensity);
} else {
// Serial.write((byte)(254 * intensity));
Serial.write((byte)(newHue));
}
}
}

if (OUTPUT_LED_DATA) {
if (DEBUG) {
Serial.println("** End FFT Results **********************");
} else {
Serial.write(255);
}
}

// pixels.show();
FastLED.show();
}

/*
//**************************************************************************
// setHSVColor
// Function - Converts HSV values to RGB and assigns the result to the specified LED
//**************************************************************************
//
void setHSVColor(int ledIndex, float hue, float saturation, float brightness) {
uint32_t myColor = pixelHSVtoRGBColor(hue, saturation, brightness);
pixels.setPixelColor(ledIndex, myColor);

// Serial.print("LED num = ");
// Serial.print(ledIndex);
// Serial.print(", color = ");
// Serial.println(myColor);
}

//**************************************************************************
// setHSVColor
// Function - Convert from HSV values to RGB colors usable by neo pixel functions.
// hue: 0.0 - 360.0
// saturation: 0.0 - 1.0
// value: 0.0 - 1.0
//**************************************************************************
//
uint32_t pixelHSVtoRGBColor(float hue, float saturation, float value) {
// Implemented from algorithm at http://en.wikipedia.org/wiki/HSL_and_HSV#From_HSV
float chroma = value * saturation;
float h1 = float(hue)/60.0;
float x = chroma*(1.0-fabs(fmod(h1, 2.0)-1.0));
float r = 0;
float g = 0;
float b = 0;
if (h1 < 1.0) {
r = chroma;
g = x;
}
else if (h1 < 2.0) {
r = x;
g = chroma;
}
else if (h1 < 3.0) {
g = chroma;
b = x;
}
else if (h1 < 4.0) {
g = x;
b = chroma;
}
else if (h1 < 5.0) {
r = x;
b = chroma;
}
else // h1 <= 6.0
{
r = chroma;
b = x;
}
float m = value - chroma;
r += m;
g += m;
b += m;
return pixels.Color(int(255*r), int(255*g), int(255*b));
}

*/

////////////////////////////////////////////////////////////////////////////////
// SPECTRUM DISPLAY FUNCTIONS
///////////////////////////////////////////////////////////////////////////////

void spectrumSetup() {
// Set the frequency window values by evenly dividing the possible frequency
// spectrum across the number of neo pixels.

// Serial.println("** spectrumSetup **********************");

float windowSize = (SAMPLE_RATE_HZ / 2.0) / float(NUM_LEDS);
for (size_t i = 0; i < ARRAY_SZ(leds) + 1; ++i) {
frequencyWindow[i] = i*windowSize;
// Serial.print("frequencyWindow[");
// Serial.print(i);
// Serial.print("] = ");
// Serial.println(frequencyWindow[i]);
}
}

//**************************************************************************
// windowMean
// Function - Compute the average magnitude of a target frequency window vs.
// all other frequencies.
// Parameters magnitudes,
// frequencyToBin(frequencyWindow[i]),
// frequencyToBin(frequencyWindow[i+1]),
// &intensity,
// &otherMean
//**************************************************************************
//
//void windowMean(float* magnitudes, uint lowBin, uint highBin, float* windowMean, float* otherMean) {
void windowMean(uint lowBin, uint highBin, float* windowMean, float* otherMean) {
*windowMean = 0;
*otherMean = 0;
// Notice the first magnitude bin is skipped because it represents the
// average power of the signal.
// Serial.print("sizeof magnitudes: ");
// Serial.print( sizeof(magnitudes) );
// Serial.print(", ARRAY_SZ(magnitudes): ");
// Serial.println( ARRAY_SZ(magnitudes) );
for (size_t i = 1; i < (ARRAY_SZ(magnitudes))/2; ++i) {

// Serial.print("i = ");
// Serial.print(i);
//
// Serial.print(", lowBin = ");
// Serial.print(lowBin);
//
// Serial.print(", highBin = ");
// Serial.print(highBin);

if (i >= lowBin && i <= highBin) {
*windowMean += magnitudes[i];
// Serial.print(", magnitude: ");
// Serial.print(magnitudes[i]);
}
else {
*otherMean += magnitudes[i];
}
// Serial.println();
}
*windowMean /= (highBin - lowBin) + 1;

// Serial.print(" **** windowMean: ");
// Serial.println(*windowMean);

*otherMean /= (FFT_SIZE / 2 - (highBin - lowBin));
}

////////////////////////////////////////////////////////////////////////////////
// SAMPLING FUNCTIONS
////////////////////////////////////////////////////////////////////////////////

//**************************************************************************
// samplingCallback
// Function - Does an analog to digital conversion of the microphone input.
//**************************************************************************
//
void samplingCallback() {

// Read from the microphone input pin and store the sample data
samples[sampleCounter] = (float32_t)analogRead(AUDIO_INPUT_PIN);

// Complex FFT functions require a coefficient for the imaginary part of the input.
// Since we only have real data, set this coefficient to zero.
samples[sampleCounter+1] = 0.0;

// Update sample buffer position and stop after the buffer is filled
sampleCounter += 2;
if (sampleCounter >= FFT_SIZE*2) {
samplingTimer.end();
}
}

//**************************************************************************
// samplingBegin
// Function - Starts the microphone ADC sampling timer function.
//**************************************************************************
//
void samplingBegin() {
// Reset sample buffer position and start callback at necessary rate.
sampleCounter = 0;
samplingTimer.begin(samplingCallback, 1000000/SAMPLE_RATE_HZ); // 1,000,000 / 9000 = 111 uSec
}

//**************************************************************************
// samplingIsDone
// Function - Indicates if a complete sample set of microphone data has been captured.
//**************************************************************************
//
boolean samplingIsDone() {
return sampleCounter >= FFT_SIZE*2;
}

////////////////////////////////////////////////////////////////////////////////
// COMMAND PARSING FUNCTIONS
// These functions allow parsing simple commands input on the serial port.
// Commands allow reading and writing variables that control the device.
//
// All commands must end with a semicolon character.
//
// Example commands are:
// GET SAMPLE_RATE_HZ;
// - Get the sample rate of the device.
// SET SAMPLE_RATE_HZ 400;
// - Set the sample rate of the device to 400 hertz.
//
////////////////////////////////////////////////////////////////////////////////

void parserLoop() {
// Process any incoming characters from the serial port
while (Serial.available() > 0) {
char c = Serial.read();
// Add any characters that aren't the end of a command (semicolon) to the input buffer.
if (c != ';') {
c = toupper(c);
strncat(commandBuffer, &c, 1);
}
else
{
// Parse the command because an end of command token was encountered.
parseCommand(commandBuffer);
// Clear the input buffer
memset(commandBuffer, 0, sizeof(commandBuffer));
}
}
}

// Macro used in parseCommand function to simplify parsing get and set commands for a variable
#define GET_AND_SET(variableName) \
else if (strcmp(command, "GET " #variableName) == 0) { \
Serial.println(variableName); \
} \
else if (strstr(command, "SET " #variableName " ") != NULL) { \
variableName = (typeof(variableName)) atof(command+(sizeof("SET " #variableName " ")-1)); \
}

//**************************************************************************
// parseCommand
// Function - Parses commands received via the serial input.
//**************************************************************************
//
void parseCommand(char* command) {
if (strcmp(command, "GET MAGNITUDES") == 0) {
for (size_t i = 0; i < ARRAY_SZ(magnitudes); ++i) {
Serial.println(magnitudes[i]);
}
}
else if (strcmp(command, "GET SAMPLES") == 0) {
for (size_t i = 0; i < ARRAY_SZ(samples); i+=2) {
Serial.println(samples[i]);
}
}
else if (strcmp(command, "GET FFT_SIZE") == 0) {
Serial.println(FFT_SIZE);
}
GET_AND_SET(SAMPLE_RATE_HZ)
// GET_AND_SET(LEDS_ENABLED)
GET_AND_SET(SPECTRUM_MIN_DB)
GET_AND_SET(SPECTRUM_MAX_DB)

// Update spectrum display values if sample rate was changed.
if (strstr(command, "SET SAMPLE_RATE_HZ ") != NULL) {
spectrumSetup();
}
}
</pre>

Machine Learning

2013-11-19T18:24:22Z

Bambikill: /* Text Stuff */

=== Join the Mailing List ===

https://www.noisebridge.net/mailman/listinfo/ml

=== Next Meeting===

*When: Tuesday, November 19, 2013 @ 7:00pm
*Where: 2169 Mission St. (Turing or Church classroom)
*Topic:
*Details: An introductory meeting. Come on out!
*Who:

=== Take the Noisebridge ML Survey ===
[http://www.surveymonkey.com/s/W2T9ZB6 Take a survey] and vote for what you want to learn!

=== Talks and Workshops ===
We've given lots of workshops and talks over the past year or so, here's a few. Many of the workshops we've given previously are recurring and will be given again, especially upon request!
*[[NBML/Workshops/Intro to Machine Learning|Intro to Machine Learning]]
*[[NBML/Workshops/Brief Tour of Statistics|A Brief Tour of Statistics]]
*[[NBML/Workshops/Generalized Linear Models|Generalized Linear Models]]
*[[NBML/Workshops/Neural Nets|Neural Nets Workshop]]
*[[NBML/Workshops/Support Vector Machines|Support Vector Machines]]
*[[NBML/Workshops/Random Forests|Random Forests]]
*[[NBML/Workshops/Independent Components Analysis|Independent Components Analysis]]
*[[NBML/Workshops/Deep Nets|Deep Nets]]

=== Code and SourceForge Site ===
*We have a [http://sourceforge.net/projects/ml-noisebridge Sourceforge Project]
*We have a git repository on the project page, accessible as:
git clone git://ml-noisebridge.git.sourceforge.net/gitroot/ml-noisebridge/ml-noisebridge
*Send an email to the list if you want to become an administrator on the site to get write access to the git repo!

=== Future Talks and Topics, Ideas ===
*Random Forests in R
*Restricted Boltzmann Machines (Mike S, some day)
*Analyzing brain cells (Mike S)
*Deep Nets w/ Stacked Autoencoders (Mike S, some day)
*Generalized Linear Models (Mike S, Erin L? some day)
*Graphical Models
*Working with the Kinect
*Computer Vision with OpenCV

=== Projects ===
*[[Small Group Subproblems]]
*[[Machine Learning/Fundraising | Fundraising]]
*[[NBML_Course|Noisebridge Machine Learning Course]]
*[[Machine Learning/Kaggle Social Network Contest | Kaggle Social Network Contest]]
*[[KDD Competition 2010]]
*[[Machine Learning/Kaggle HIV | HIV]]

=== [[Machine_Learning/Datasets|Datasets and Websites]] ===
*[http://archive.ics.uci.edu/ml/ UCI Machine Learning Repository]
*[[DataSF.org]]
*[http://infochimps.com/ Infochimps]
*[http://www.face-rec.org/databases/ Face Recognition Databases]
*[http://robjhyndman.com/TSDL/ Time Series Data Library]
*[http://getthedata.org/ Data Q&A Forum]
*[http://metaoptimize.com/qa/ Metaoptimize]
*[http://www.quora.com/Machine-Learning Quora ML Page]
*[http://www.metoffice.gov.uk/research/climate/climate-monitoring/land-and-atmosphere/surface-station-records A ton of Weather Data]
*[http://mlcomp.org/ MLcomp]
**Upload your algorithm and objectively compare it's performance to other algorithms
*[http://www.ntis.gov/products/ssa-dmf.aspx Social Security Death Master File!]

=== Software Tools ===

==== Generic ML Libraries ====
*[http://www.cs.waikato.ac.nz/ml/weka/ Weka]
**a collection of data mining tools and machine learning algorithms.
*[http://scikit-learn.sourceforge.net/ scikits.learn]
**Machine learning Python package
*[http://pypi.python.org/pypi/scikits.statsmodels scikits.statsmodels]
**Statistical models to go with scipy
*[http://pybrain.org PyBrain]
**Does feedforward, recurrent, SOM, deep belief nets.
*[http://www.csie.ntu.edu.tw/~cjlin/libsvm/ LIBSVM]
**c-based SVM package
*[http://pyml.sourceforge.net PyML]
*[http://mdp-toolkit.sourceforge.net/ MDP]
**Modular framework, has lots of stuff!
*[[Machine Learning/VirtualBox|VirtualBox]] Virtual Box Image with Pre-installed Libraries listed here
*[http://deeplearning.net/software/theano/ Theano: Symbolic Expressions and Transparent GPU Integration]
*[http://sympy.org sympy] Does symbolic math
*[http://waffles.sourceforge.net/ Waffles]
**Open source C++ set of machine learning command line tools.
*[http://rapid-i.com/content/view/181/196/ RapidMiner]
*[http://www.mrpt.org/ Mobile Robotic Programming Toolkit]
*[http://nipy.sourceforge.net/nitime/ nitime]
**NeuroImaging in Python, has some good time series analysis stuff and multi-variate response fitting.
*[http://pandas.pydata.org/ Pandas]
**Data analysis workflow in python
*[http://www.pytables.org/moin PyTables]
**Adds querying capabilities to HDF5 files
*[http://statsmodels.sourceforge.net/ statsmodels]
**Regression, time series analysis, statistics stuff for python
*[https://github.com/JohnLangford/vowpal_wabbit/wiki Vowpal Wabbit]
**"Intrinsically Fast" implementation of gradient descent for large datasets
*[http://www.shogun-toolbox.org/ Shogun]
**Fast implementations of SVMs
*[http://www.mlpack.org/ MLPACK]
**High performance scalable ML Library
*[http://www.torch.ch/ Torch]
**MATLAB-like environment for state-of-the art ML libraries written in LUA

==== Online ML ====
*[http://moa.cs.waikato.ac.nz/ MOA (Massive Online Analysis)]
**Offshoot of weka, has all online-algorithms
*[http://jubat.us/en/ Jubatus]
**Distributed Online ML
*[http://dogma.sourceforge.net/ DOGMA]
**MATLAB-based online learning stuff
*[http://code.google.com/p/libol/ libol]
*[http://code.google.com/p/oll/ oll]
*[http://code.google.com/p/scw-learning/ scw-learning]

==== Graphical Models ====
*[http://www.mrc-bsu.cam.ac.uk/bugs/ BUGS]
**MCMC for Bayesian Models
*[http://mcmc-jags.sourceforge.net/ JAGS]
**Hierarchical Bayesian Models
*[http://mc-stan.org/ Stan]
**A graphical model compiler
*[https://github.com/kutschkem/Jayes Jayes]
**Bayesian networks in Java

==== Text Stuff ====
*[http://www.crummy.com/software/BeautifulSoup/ Beautiful Soup]
**Screen-scraping tools
*[http://www.mlsec.org/sally/ SALLY]
**Tool for embedding strings into vector spaces
*[http://radimrehurek.com/gensim/ Gensim]
**Topic modeling

==== Collaborative Filtering ====
*[http://prea.gatech.edu/ PREA]
**Personalized Recommendation Algorithms Toolkit
*[http://svdfeature.apexlab.org/wiki/Main_Page SVDFeature]
**Collaborative Filtering and Ranking Toolkit

==== Computer Vision ====
*[http://opencv.willowgarage.com/documentation/index.html OpenCV]
**Computer Vision Library
**Has ML component (SVM, trees, etc)
**Online tutorials [http://www.pages.drexel.edu/~nk752/tutorials.html here]
*[http://drwn.anu.edu.au/ DARWIN]
**Generic C++ ML and Computer Vision Library
*[http://sourceforge.net/projects/petavision/ PetaVision]
**Developing a real-time, full-scale model of the primate visual cortex.

==== Audio Processing ====
*[http://tlecomte.github.com/friture/ Friture]
**Real-time spectrogram generation
*[http://code.google.com/p/pyo/ pyo]
**Real-time audio signal processing
*[https://github.com/jsawruk/pymir PYMir]
**A library for reading mp3's into python, and doing analysis
*[http://www.fon.hum.uva.nl/praat/ PRAAT]
**Speech analysis toolkit
*[http://ofer.sci.ccny.cuny.edu/sound_analysis_pro Sound Analysis Pro]
**Tool for analyzing animal sounds
*[http://luscinia.sourceforge.net/ Luscinia]
**Software for archiving, measuring, and analyzing bioacoustic data

*[http://wiki.python.org/moin/PythonInMusic List of Sound Tools for Python]

==== Data Visualization ====
*[http://www.ailab.si/orange/ Orange]
**Strong data visualization component
*[http://gephi.org/ Gephi]
**Graph Visualization
*[http://had.co.nz/ggplot2/ ggplot]
**Nice plotting package for R
*[http://code.enthought.com/projects/mayavi/ MayaVi2]
**3D Scientific Data Visualization
*[http://cytoscape.github.io/cytoscape.js/ Cytoscape]
**A JavaScript graph library for analysis and visualisation
*[https://plot.ly/ plot.ly]
**Web-based plotting

==== Cluster Computing ====
*[http://lucene.apache.org/mahout/ Mahout]
**Hadoop cluster based ML package.
*[http://web.mit.edu/star/cluster/ STAR: Cluster]
**Easily build your own Python computing cluster on Amazon EC2

==== Database Stuff ====
*[http://madlib.net/ MADlib]
**Machine learning algorithms for in-database data
*[http://www.joyent.com/products/manta Manta]
**Distributed object storage

==== Neural Simulation ====
*[http://nengo.ca/ Nengo]

==== Other ====
*[http://jmlr.csail.mit.edu/mloss/ Journal of Machine Learning Software List]

=== Presentations and other Materials ===
* [[Awesome Machine Learning Applications]] -- A list of cool applications of ML
* [[Hands-on Machine Learning]], a presentation [[User:jbm|jbm]] gave on 2009-01-07.
* http://www.youtube.com/user/StanfordUniversity#g/c/A89DCFA6ADACE599 Stanford Machine Learning online course videos]
* [[Media:Brief_statistics_slides.pdf]], a presentation given on statistics for the machine learning group
* [http://www.linkedin.com/groupAnswers?viewQuestionAndAnswers=&discussionID=20096092&gid=77616&trk=EML_anet_qa_ttle-0Nt79xs2RVr6JBpnsJt7dBpSBA LinkedIn] discussion on good resources for data mining and predictive analytics
* [http://www.face-rec.org/algorithms/ Face Recognition Algorithms]
* [http://www.ics.uci.edu/~welling/classnotes/classnotes.html Max Welling's ML classnotes]

=== Topics to Learn and Teach ===
[[NBML Course]] - Noisebridge Machine Learning Curriculum (work-in-progress)

[[CS229]] - The Stanford Machine learning Course @ noisebridge

*Supervised Learning
**Linear Regression
**Linear Discriminants
**Neural Nets/Radial Basis Functions
**Support Vector Machines
**Classifier Combination [http://www.cedar.buffalo.edu/~srihari/CSE555/Chap9.Part6.pdf]
**A basic decision tree builder, recursive and using entropy metrics

*Unsupervised Learning
**[[Machine_Learning/HMM|Hidden Markov Models]]
**Clustering: PCA, k-Means, Expectation-Maximization
**Graphical Modeling
**Generative Models: gaussian distribution, multinomial distributions, HMMs, Naive Bayes
**[[Machine_Learning/Deep_Belief_Networks|Deep Belief Networks & Restricted Boltzmann Machines]]

*Reinforcement Learning
**Temporal Difference Learning

*Math, Probability & Statistics
**Metric spaces and what they mean
**Fundamentals of probabilities
**Decision Theory (Bayesian)
**Maximum Likelihood
**Bias/Variance Tradeoff, VC Dimension
**Bagging, Bootstrap, Jacknife [http://www.cedar.buffalo.edu/~srihari/CSE555/Chap9.Part3.pdf]
**Information Theory: Entropy, Mutual Information, Gaussian Channels
**Estimation of Misclassification [http://www.cedar.buffalo.edu/~srihari/CSE555/Chap9.Part5.pdf]
**No-Free Lunch Theorem [http://www.cedar.buffalo.edu/~srihari/CSE555/Chap9.Part1.pdf]

*Machine Learning SDK's
** [http://opencv.willowgarage.com/documentation/index.html OpenCV] ML component (SVM, trees, etc)
**[http://lucene.apache.org/mahout/ Mahout] a Hadoop cluster based ML package.
**[http://www.cs.waikato.ac.nz/ml/weka/ Weka] a collection of data mining tools and machine learning algorithms.

*Applications
** Collective Intelligence & Recommendation Engines

=== [[Machine Learning/Meeting Notes|Meeting Notes]]===

[[Category:Events]]
[[Category:Projects]]