Principles of Deep Learning
Principles of Deep Learning
Contents Hours
Unit 1 Intro to neural networks cost functions, hypotheses and tasks; training data;
maximum likelihood, based cost, cross entropy, MSE cost; feed-forward
8
networks; MLP, sigmoid units; neuroscience, inspiration; Learning in neural
networks output vs hidden layers; linear vs nonlinear networks.
Unit 2 Backpropagation learning via gradient descent; recursive chain rule
(backpropagation); if time:Bias-variance tradeoff, regularization; output units: 12
linear, soft max; hidden.
Unit 3 Deep learning strategies I (e.g., GPU training, regularization,etc); project
proposals, Deep learning strategies II (e.g., RLUs, dropout, etc) SCC/TensorFlow
8
overview How to use theSCC cluster; introduction to Tensorflow. CNNs I
Convolution neural networks.
Unit 4 Deep Belief Nets I probabilistic methods RNNs I Recurrent neural networks
Other DNN variants(e.g. attention, memory networks, etc.), Neural Turing 8
Machines(Kate)
Unit 5 Unsupervised deep learning
Unsupervised deep learningI, Unsupervised deep learning II (e.g. deep
9
generative models etc.), Deep reinforcement learning Vision applications I NLP
applications I Laboratory Exercises
Suggested Readings:
1. Deep Learning and Artificial Intelligence: A Beginners' Guide to Neural Networks and Deep Learning, By
John Slavio
2. Introduction to Deep Learning: From Logical Calculus to Artificial Intelligence By Sandro Skansi
3. MATLAB Deep Learning: With Machine Learning, Neural Networks and Artificial, By Phil Kim