Machine Learning Guide

This series aims to teach you the high level fundamentals of machine learning from A to Z. I'll teach you the basic intuition, algorithms, and math. We'll discuss languages and frameworks, deep learning, and more. Audio may be an inferior medium to task; but with all our exercise, commute, and chores hours of the day, not having an audio supplementary education would be a missed opportunity. And where your other resources will provide you the machine learning trees, I’ll provide the forest. Additionally, consider me your syllabus. At the end of every episode I’ll provide the best-of-the-best resources curated from around the web for you to learn each episode’s details.

Machine Learning Guide is a how to podcast hosted by OCDevel, with 30 episodes on record and a Required Pod Score of 80. PitchCentric scores this show on Booking Probability, Listen Score, and live audience signals refreshed every 24 hours.

OCDevel hosts Machine Learning Guide, a how to show with 30 episodes published.

Our AI reads these to draft pitches. Use them as grounding for a pitch that cites a real guest and a specific topic.

• 

  Machine Learning Guide

  30. New Series: Machine Learning Applied

  May 24, 2018—

  Introducing a new podcast series on Patreon: Machine Learning Applied MLG: I'm rebooting this series to fix mistakes & add more shallows (Bayesian methods, Tree methods, etc). I'm adding Patreon rewards, including access to a new podcast series: Machine Learning Applied, discussing applied/practical 10-20m frequent episodes.

  ListenShow Notes
• 

  Machine Learning Guide

  29. Reinforcement Learning Intro

  Feb 4, 2018—

  Introduction to reinforcement learning concepts ## Resources - Hands-On Machine Learning with Scikit-Learn and TensorFlow (https://www.amazon.com/dp/1491962291?tag=ha0d2-20) `book:medium` (last chapter) - undefined - AI a Modern Approach. Website (http://aima.cs.berkeley.edu/), Book (https://www.amazon.com/dp/0136042597?tag=ha0d2-20) `book:hard` - Berkeley cs294: Deep Reinforcement Learning (http://rll.berkeley.edu/deeprlcourse/) `course:hard` - RL Course by David Silver (https://www.youtube.com/playlist?list=PLzuuYNsE1EZAXYR4FJ75jcJseBmo4KQ9-) `course|audio:hard` - Convert video to audio: ** mp4 => mp3: `for f in *.mp4; do ffmpeg -i "$f" "${f%.mp4}.mp3" && rm "$f"; done` ** youtube => mp3: setup youtube-dl (https://github.com/rg3/youtube-dl) and run `youtube-dl -x youtube.com/playlist?list= ` ## Episode - RL definition: goal, rewards, actions ** Games (Atari, Chess, Go - Lee Sedol & Alpha Go) ** AI: learning, vision / speech, action / motion, planning ** Reasoning / knowledge vs model-based Deep RL? ** Reasoning / knowledge rep (+memory?) => Differential computers (https://deepmind.com/blog/differentiable-neural-computers/) ** vs supervised. Vision = supervised. Games = action. Trading can go both ways! ** Time: Credit assignment, delayed rewards, investment - Model-based v free ** Policy (what you do; gut reaction) - Value-based (Q-learning) vs Policy Gradient ** PG is direct: ML -> action ** Value-based indirect: Bellman stuff -> state/action values (Q-values) -> policy - Openai Gym, cartpole - Frameworks ** openai/baselines (https://github.com/openai/baselines) ** reinforceio/tensorforce (https://github.com/reinforceio/tensorforce) ** NervanaSystems/coach (https://github.com/NervanaSystems/coach) ** rll/rllab (https://github.com/rll/rllab)

  ListenShow Notes
• 

  Machine Learning Guide

  28. Hyperparameters 2

  Feb 3, 2018—

  Hyperparameters part 2: hyper-search, regularization, SGD optimizers, scaling ## Episode - Hyper optimization ** GridSearch, RandomSearch ** Bayesian Optimization (https://thuijskens.github.io/2016/12/29/bayesian-optimisation/) - Regularization: Dropout, L2, L1 ** DNNs = Dropout ** L2 = most common ** L1 = sparsity (zeros) & feature-selection (rarer circumstances) - Optimizers (SGD): Momentum -> Adagrad -> RMSProp -> Adam -> Nadam (http://sebastianruder.com/optimizing-gradient-descent/index.html#visualizationofalgorithms) - Initializers: Zeros, Random Uniform, Xavier - Scaling ** Feature-scaling: MinMaxScaler, StandardScaler, RobustScaler ** Features + inter-layer: Batch Normalization

  ListenShow Notes
• 

  Machine Learning Guide

  27. Hyperparameters 1

  Jan 26, 2018—

  Hyperparameters part 1: network architecture ## Episode - Hypers future & meta-learning ** We're always removing hypers. DL removed feature-engineering ` - Model selection ** Unsupervised? K-means Clustering => DL ** Linear? Linear regression, logistic regression ** Simple? Naive Bayes, Decision Tree (Random Forest, Gradient Boosting) ** Little data? Boosting ** Lots of data, complex situation? Deep learning - Network ** Layer arch ** Vision? CNN ** Time? LSTM ** Other? MLP ** Trading LSTM => CNN decision ** Layer size design (funnel, etc) ** Face pics ** From BTC episode ** Don't know? Layers=1, Neurons=mean(inputs, output) https://stats.stackexchange.com/a/1097/107199 - Activations / nonlinearity https://towardsdatascience.com/activation-functions-neural-networks-1cbd9f8d91d6 ** Output ** Sigmoid = predict probability of output, usually at output ** Softmax = multi-class ** Nothing = regression ** Relu family (Leaky Relu, Elu, Selu, ...) = vanishing gradient (gradient is constant), performance, usually better ** Tanh = classification between two classes, mean 0 important

  ListenShow Notes
• 

  Machine Learning Guide

  26. Project Bitcoin Trader

  Jan 25, 2018—

  Community project & intro to Bitcoin/crypto + trading ## Resources - TForce BTC Trader (https://github.com/lefnire/tforce_btc_trader) (podcast project) ## Episode - Project: Trading Crypto ** Reasons ** Get rich ** Hot topic ** Special: Intuitively highlights decisions: hypers, supervised v reinforcement, LSTM v CNN ** Pros: skip this - Crypto (v stock) ** Bitcoin, Ethereum, Litecoin, Ripple ** Many benefits (immutable permenant distributed ledger; security; low fees; international; etc) ** For our purposes: popular, volatile, singular ** Singular like Forex vs Stock (instruments) - Trading basics ** Day, swing, investing ** Patterns (technical analysis, vs fundamentals) ** OHLCV / Candles ** Indicators ** Exchanges & Arbitrage (GDAX, Krakken) - Good because highlights lots ** LSTM v CNN ** Supervised v Reinforcement ** Obvious net architectures (indicators, time-series, tanh v relu)

  ListenShow Notes

Based on semantic analysis of episode topics and host coverage, this show is a strong guest fit for executives in:

Industry fit is computed by PitchCentric using vector embeddings of the show's episode catalog.