L10a Representation learning

Lecture slides (Lect10a Download Lect10a)

Date and venue: Feb 19, kl.15:15-17:00, B2

Session format: Building on the pre-recorded introduction to deep learning fundamentals (L9) and representation learning, a more detailed account of selected aspects will be given in an online seminar format combined with group discussions and Q&A

Before the lecture

Pre-recorded video on representation learning and lab 4 DBN 

Please watch also pre-recorded material for lecture L9. 

Discussion Download Discussion accompanying Lecture10a.

After the lecture
Slides Download Slides , Recorded Zoom video

  

Scope, topics covered

  • Generative models
  • Restricted Boltzmann machines (RBMs), contrastive divergence
  • Gaussian Bernoulli RBMs for continuous data
  • Autoencoders
  • Data representations

 

Reading material:

  • I. Goodfellow et al.: ch.14, 15, 18 (18.1-18.2), 20 (20.1-20.5)

  • tutorials/survey papers, e.g.:
    • Salakhutdinov, R. (2015) Learning deep generative models. Annual Reviews of Statistics and Its Application, 2, p.361–385.
    • Bengio, Y., Courville, A., & Vincent, P. (2013). Representation learning: A review and new perspectives. IEEE transactions on pattern analysis and machine intelligence, 35 (8), p.1798-1828.
    • Bengio, Y. (2009) Learning deep architectures for AI. Foundations and trends® in Machine Learning, 2.1. p.1-127.