![](http://the-charlie.com/ml/wp-content/uploads/2017/12/1cuTSPlTq0a_327iTPJyD-Q.png)
Image care of https://towardsdatascience.com/the-mostly-complete-chart-of-neural-networks-explained-3fb6f2367464
Learning Resources for Artificial Intelligence
Math
- Linear Algebra
- Essence of Linear Algebra by 3blue1brown: https://www.youtube.com/playlist?list=PLZHQObOWTQDPD3MizzM2xVFitgF8hE_ab
- Gilbert Strang’s famous intro to Linear Algebra: https://ocw.mit.edu/courses/mathematics/18-06-linear-algebra-spring-2010/video-lectures/
- Essence of Calculus: https://www.youtube.com/playlist?list=PLZHQObOWTQDMsr9K-rj53DwVRMYO3t5Yr
- Probability:
- Jaynes : https://www.amazon.com/Probability-Theory-Science-T-Jaynes/dp/0521592712
- Rethinking Probability : http://xcelab.net/rm/statistical-rethinking/
- Differential Equations: https://ocw.mit.edu/resources/res-18-009-learn-differential-equations-up-close-with-gilbert-strang-and-cleve-moler-fall-2015/index.htm
- Real Analysis: https://www.youtube.com/watch?v=sqEyWLGvvdw
- Seeing Theory: http://students.brown.edu/seeing-theory/
Traditional AI
These topics fall outside of ‘Machine Learning’ or ‘Deep Learning’
- AI: A modern approach
- The classic textbook and the 4th most cited book in the 21st century.
- The cheap copy of the book : https://www.amazon.com/Artificial-Intelligence-Approach-Stuart-Russell/dp/9332543518/
- Exercises and Notebooks :https://github.com/aimacode
- Game AI
- AlphaBeta demo from Berkeley – https://www.youtube.com/watch?v=xBXHtz4Gbdo
- Tic Tac Toe with MiniMax – https://medium.freecodecamp.com/how-to-make-your-tic-tac-toe-game-unbeatable-by-using-the-minimax-algorithm-9d690bad4b37#.sr2qz5bhq
- AlphaBeta Pruning – https://www.youtube.com/watch?v=xBXHtz4Gbdo
- Search AI
- Heuristics (l8 and l9) – https://heuristicswiki.wikispaces.com/
- Sebastian Thrun Q&A – https://www.youtube.com/watch?v=AijSw9Q3oas
- Search Heuristic (l7 – Pacman) – http://theory.stanford.edu/~amitp/GameProgramming/Heuristics.html
- Logic
- Description Logics – http://www.cs.ox.ac.uk/people/ian.horrocks/Publications/download/2007/BaHS07a.pdf
- Knowledge Representation & Classical Logic – http://www.cs.utexas.edu/~vl/papers/Ch1.pdf
- First-Order Logic – https://formal.iti.kit.edu/~beckert/teaching/Einfuehrung-KI-WS0304/08FirstOrderLogic.pdf
- Overview of AI Planning – https://users.ics.aalto.fi/rintanen/planning.html
- Bayes Nets
- Hidden Markov Models
- HMM for DNA Analysis – http://www.lancs.ac.uk/~nemeth/Hidden%20Markov%20Models%20with%20Applications%20to%20DNA%20Sequence%20Analysis.pdf
- HMM for Speech Recognition – http://www.cslu.ogi.edu/~zak/cs506-lvr/mjfg_NOW.pdf
- HMM for Bioinformatics – http://www.cs.ubbcluj.ro/~csatol/mach_learn/bemutato/Mate_Korosi_HMMpres.pdf
Machine Learning
Come back to this, include things like SVM, k-NN, Naive Bayes, Random Forests, Linear Regression, LASSO, Decision Trees, Feature Learning, PCA .
- A really good look at linear regression in R : https://rpubs.com/ronniekolodziej/156135
- R for data science : http://r4ds.had.co.nz/
- Applied Predictive Modelling
Deep Learning
( Incomplete, just filling this in so I dont lose the RNN links )
- Recurrent Neural Networks
- RNN in 100 lines of python https://gist.github.com/karpathy/d4dee566867f8291f086
- http://karpathy.github.io/2015/05/21/rnn-effectiveness/
- http://colah.github.io/posts/2015-08-Understanding-LSTMs/
- For Speech Recognition https://svds.com/tensorflow-rnn-tutorial/
- CNN
- Reinforcement Learning
- Anything from Andrej Karpathy is worth reading – http://karpathy.github.io/2016/05/31/rl/
- Policy Gradients: http://pemami4911.github.io/blog/2016/08/21/ddpg-rl.html
- Notebooks and Exercises: https://github.com/dennybritz/reinforcement-learning
- NLP
- Stanford taught by Richard Socher https://www.youtube.com/watch?v=Qy0oEkCZkBI&list=PLlJy-eBtNFt4CSVWYqscHDdP58M3zFHIG&index=1
- Speech Recognition
- Feature Extraction : http://www.ijcsmc.com/docs/papers/March2015/V4I3201545.pdf
- python_speech_features: https://github.com/jameslyons/python_speech_features
- Acoustic Model Features: https://pdfs.semanticscholar.org/a566/cd4a8623d661a4931814d9dffc72ecbf63c4.pdf
- Ceptral Analysis : http://iitg.vlab.co.in/?sub=59&brch=164&sim=615&cnt=1
- MFCC Tutorial : http://practicalcryptography.com/miscellaneous/machine-learning/guide-mel-frequency-cepstral-coefficients-mfccs/
- Connectionist Temporal Classification http://machinelearning.wustl.edu/mlpapers/paper_files/icml2006_GravesFGS06.pdf
- Audio Analysis with TensorFlow: https://medium.com/towards-data-science/audio-processing-in-tensorflow-208f1a4103aa
- https://people.xiph.org/~jm/demo/rnnoise/
GPU enabled Ubuntu
After botching the install 4 or 5 times I finally found something that worked.
- Install Ubuntu LTS fresh
- Install NVIDIA drivers FROM THE UBUNTU REPOSITORY ( Usually named “Additional Drivers” ) .
- Install the CUDA libraries from NVIDIA , https://developer.nvidia.com/cuda-downloads. Choose runfile(local). When prompted, DO NOT LET NVIDIA OVERWRITE THE DISPLAY DRIVER.
- Install cudacdnn from https://developer.nvidia.com/cudnn , you have to register unfortunately.
- Install conda from https://www.continuum.io/downloads
pip install keras theano tensorflow-gpu
- To force Tensorflow into using a specific device
-
with tf.device('gpu:0'): # train your model