This schedule is tentative and often updated. Please access this page regularly to have the most recent materials. Timely announcements will be made on Piazza as well.
Lecture | Date | Description | Materials |
---|---|---|---|
Lecture 1 | Tue Jan 14 |
Course Introduction
Course logistics ML Reviews |
[slides1]
[slides2] |
Lecture 2 | Thu Jan 16 |
ML Reviews
Error and Loss Approximate Optimization |
[slides1]
[slides2] |
Lecture 3 | Thu Jan 21 |
ML Reviews
ML Models Neural Networks |
[slides] |
SUPERVISED LEARNING |
|||
Lecture 4 | Tue Jan 23 |
Feedforward Neural Networks
Training Feedforward NNs Tensorflow and Variables Building NNs Training with Tensorflow |
[slides]
|
Lecture 5 | Thu Jan 28 |
Convolution
Convolution Operations Case Study |
[slides]
|
Lecture 6 | Tue Jan 30 |
Convolution (cont'd)
Case Study Visualization Graph Convolution |
[slides]
|
Assignment 1 | Tue Feb 04 |
Due: Feb 23 | [link] |
LAB | Tue Feb 04 |
Compute Resources for Deep Learning
Amazon AWS |
[slides]
|
Lecture 7 | Thu Feb 06 |
Customizing NNs
Input data Auto Differentiation Losses Metrics |
[slides]
|
Lecture 8 | Tue Feb 11 |
Customizing NNs
Metrics Optimizers |
[slides]
|
Lecture 9 | Tue Feb 13 |
Customizing NNs
Optimizers Activations Callbacks |
[slides]
|
Lecture 10 | Tue Feb 18 |
Customizing NNs
Callbacks Save and Restore Tranfer Learning |
[slides]
|
Lecture 11 | Thu Feb 20 |
Regularization and Other Training Techniques
Data Cleansing Overfitting Batchnorm |
[slides]
|
Lecture 12 | Thu Feb 25 |
Regularization and Other Training Techniques
Initialization Regularization Dropout |
[slides]
|
Lecture 13 | Tue Feb 27 |
Recurrent Neural Network
Composition of Functions Modelling RNNs Extensions |
[slides]
|
Assignment 2 | Tue Mar 01 |
Due: Mar 27 | [link] |
Lecture 14 | Thu Mar 03 |
Recurrent Neural Networks (cont'd)
NMT: Training |
[slides]
|
Lecture 15 | Thu Mar 05 |
Recurrent Neural Networks (cont'd)
Attention NMT: Decoding Regularization |
[slides]
|
Spring Break - No classes | Tue Mar 10 |
||
Spring Break - No classes | Thu Mar 12 |
||
Spring Break (extended) - No classes | Tue Mar 17 |
||
Lecture 16 | Thu Mar 19 |
Recurrent Neural Networks (cont'd)
NMT: Decoding Regularization |
[slides]
|
LARGE-SCALE MACHINE LEARNING |
|||
Lecture 17 | Tue Mar 24 |
Distributed Tensorflow
Managing Devices Multiple GPUs Multiple Machines |
[slides]
|
Lecture 18 | Tue Mar 26 |
Distributed Pytorch
Brief Pytorch Introduction Data Parallelism Pytorch Distributed |
[slides]
|
UNSUPERVISED LEARNING |
|||
Lecture 19 | Tue Mar 31 |
Self-Supervised Learning Introduction
Data and Methodology (Auto)Encoder-Decoder Encoding Decoding |
[slides]
|
Lecture 20 | Thu Apr 2 |
Self-Supervised Learning Introduction (cont'd)
Encoding Decoding |
[slides]
|
Assignment 3 | Tue Apr 05 |
Due: Apr 25 | [link] |
Lecture 21 | Tue Apr 7 |
VAE
Variational Inference ELBO AutoEncoder to VAE Reparameterization |
[slides]
|
Lecture 22 | Tue Apr 9 |
VAE Implementation
Encoder Decoder Prior Posterior Collapse |
[slides]
|
Lecture 23 | Tue Mar Apr 14 |
Generative Adversarial Networks
Motivation from AE Minimax GAN |
[slides]
|
Lecture 24 | Thu Apr 16 |
GAN Implementation
MNIST Sample Training Problems Architectures |
[slides]
|
Lecture 25 | Thu Apr 21 |
GAN Implementation
Losses Evaluation |
[slides]
|
Lecture 25 | Thu Apr 23 |
Conditional Generation
CVAE, CGAN Image Style Transfer |
[slides]
|
Lecture 26 | Thu Apr 28 |
Conditional Generation (cont'd)
Text Style Transfer Cross-Domain Transfer |
[slides]
|
Lecture 27 | Thu Apr 30 |
Normalizing Flows
Transformation of variables Residual Flows Autoregressive Flows Discrete Flows |
[slides]
|