Upgrade to Pro — share decks privately, control downloads, hide ads and more …

ML Session n°6

ML Session n°6


Adrien Couque

May 16, 2017


  1. ML: recurrent neural networks April 2017

  2. Recap

  3. Dropout

  4. Fully connected vs Convolution

  5. “Deep” layers

  6. Convolutional architecture

  7. Convolutional network -> fully connected

  8. Sliding window with different scales

  9. Transforming a panda into a vulture

  10. RNN

  11. Limits of DNN/CNN - can only handle fixed-size inputs -

    unable to handle temporal data - no memory
  12. New brick

  13. Unrolling an RNN

  14. Predicting the next letter

  15. Example: translation

  16. Example: speech-to-text

  17. Multiple layers

  18. Different architectures

  19. Training RNN: Backpropagation through time

  20. Problem: long term dependencies

  21. Problem: long term dependencies

  22. LSTM

  23. Inside a standard RNN cell

  24. Inside an LSTM cell

  25. Decomposition: cell inner state

  26. Forget gate

  27. Update gate

  28. Updating the state

  29. Calculating an output

  30. GRU

  31. GRU (Gated Recurrent Unit)

  32. Demo !

  33. Example : Wikipedia Naturalism and decision for the majority of

    Arab countries' capitalide was grounded by the Irish language by [[John Clair]], [[An Imperial Japanese Revolt]], associated with Guangzham's sovereignty. His generals were the powerful ruler of the Portugal in the [[Protestant Immineners]], which could be said to be directly in Cantonese Communication, which followed a ceremony and set inspired prison, training. The emperor travelled back to [[Antioch, Perth, October 25|21]] to note, the Kingdom of Costa Rica, unsuccessful fashioned the [[Thrales]], [[Cynth's Dajoard]], known in western [[Scotland]], near Italy to the conquest of India with the conflict. Copyright was the succession of independence in the slop of Syrian influence that was a famous German movement based on a more popular servicious, non-doctrinal and sexual power post. Many governments recognize the military housing of the [[Civil Liberalization and Infantry Resolution 265 National Party in Hungary]], that is sympathetic to be to the [[Punjab Resolution]] (PJS)[http://www.humah.yahoo.com/guardian. cfm/7754800786d17551963s89.htm Official economics Adjoint for the Nazism, Montgomery was swear to advance to the resources for those Socialism's rule, was starting to signing a major tripad of aid exile.]]
  34. Example : LATEX

  35. Word2Vec

  36. Word embedding?

  37. Word2Vec

  38. Implementing word2vec

  39. Implementing word2vec

  40. Yelp dataset

  41. Yelp dataset

  42. Next Unsupervised Learning Reinforcement Learning April 2017