L9 Deep learning fundamentals: general philosophy and a review of deep architectures

Lecture slides (Lect9 Download Lect9)

Date and venue: Oct 3, kl.8:15-10:00, Lecture hall K1

Session format:
A summary of the pre-recorded material for deep learning (live lecture format) combined with group discussions and Q&A 

Before the lecture

(intro to Lecture 9) Pre-recorded video on introduction to deep learning: pre-recorded video, Download slides

 

After the lecture
Recorded Zoom video - interactive session with mentimeter and discussion including students

  

Scope, topics covered

  • The concept of deep learning / deep architectures
  • Key architectures and algorithms
  • General functionality and new opportunities
  • Scope of applications
  • Fundamental hypotheses

 

Reading material:

  • I. Goodfellow et al.: ch. 6, 9, 12

  • tutorials/survey papers available on deeplearning.net, e.g.:
    • LeCun, Y., Bengio, Y., & Hinton, G. (2015) Deep learning. Nature, 521, p.436–444.
    • Schmidhuber, J. (2015). Deep Learning and Neural Networks: An Overview. Neural Networks, 61, p.85-117.
    • Bengio, Y. (2009). Learning deep architectures for AI. Foundations and Trends in Machine Learning, 2(1), pp. 1-127.