Do Androids Dream of Electronic Dance Music?

Slides for my RubyKaigi 2017 talk on machine learning + music, co-presented with the inimitable Julian Cheal!

Transcript

1. Do Androids Dream of Electronic Dance Music? Julian Cheal and Eric Weinstein # Ruby Kaigi 2017 # Hiroshima, Japan # 18 September 2017
2. Do Androids What now?
3. Part 0: Hello! こんにちは!
4. Who Are We?
5. Julian Cheal Rubyist RedHat
6. Bath
9. About Eric eric_weinstein = { # エリック employer: 'Fox Networks Group', github: 'ericqweinstein', twitter: 'ericqweinstein', website: 'ericweinste.in' } 30% off with KAIGI30!
10. Agenda • Introduction to machine learning • Neural networks, RNNs, and LSTM RNNs • Midi • Demo! Synths!? • Questions?
11. Part 1: Machine Learning
12. bit.ly/EricsMLTalk
13. What Is It? • Machine learning is generalization • Supervised learning is generalizing from
 labeled data to unlabeled data
14. Supervised Learning • Classification or regression • Learning patterns from labeled data, then generalizing to unlabeled data
15. Part 2: Neural Networks
16. • An artificial neural network is a machine learning model that mimics the structure and function of a biological brain • In our case, we’re interested in a particular kind of recurrent neural network:
 LSTM (long/short-term memory) network What Is It?
17. Image credit: https://github.com/cdipaolo/goml/tree/master/perceptron
18. Image credit: https://en.wikipedia.org/wiki/Artificial_neural_network
19. Image credit: https://leonardoaraujosantos.gitbooks.io/artificial- inteligence/content/recurrent_neural_networks.html
20. Part 3: Midi
21. bit.ly/JuliansMusicTalk
22. What is Midi? Dave Smith “Father of Midi” Midi invented in 1983
23. Midi Note Data
24. Midi Note Data
25. Find some Data
26. deadmau5 - Chimaera https://splice.com/sounds/ splice/deadmau5-chimaera
27. Convert the Data Midi TFRecord
28. Convert the Data
29. Moar Data
30. Bring your Own Data
32. Show me the Code
33. Part 4: Dance Party
34. Overview • The technology (Python, Ruby, Tensorflow, Magenta) • The architecture of our neural network • How did training go? • Demo time/dance party!
35. We need More Ruby! • Python 93.4% • Ruby 4.8% • Shell 1.8% WTF! This is Ruby Kaigi
37. Summary • Machine learning is generalization • Neural networks are machine learning models based on biological brains • RNNs are a kind of NN with “loop”; feedback allows modeling for text prediction, music generation, and more • We can do this all with Ruby, we just need the library support!
38. 3.times { puts 'ありがとう!' }
39. Questions? @ericqweinstein @juliancheal