13 – Self-Supervised Learning (30/03/2022-06/04/2022)

In this lecture, we will discuss self-supervised learning. We will discuss how to create representation beyond the supervised pre-training paradigm, and we are going to see how effective pretext tasks can be designed and how to train with contrastive objectives.

Slides:

• SSL - pretext tasks slides

• SSL - contrastive methods (plus bonus material at the end)

Video of SSL lecture I. (30/03/2022)
Video of SSL lecture II. (04/04/2022)
Video of SSL lecture III. (06/04/2022)

Reference:

• Doersch, Carl, Abhinav Gupta, and Alexei A. Efros. "Unsupervised visual representation learning by context prediction." CVPR (2015).

• Gidaris, Spyros, Praveer Singh, and Nikos Komodakis. "Unsupervised representation learning by predicting image rotations." ICLR (2018).

• Wu, Zhirong, et al. "Unsupervised feature learning via non-parametric instance discrimination." CVPR (2018).

• He, Kaiming, et al. "Momentum contrast for unsupervised visual representation learning." CVPR (2020).

• Chen, Ting, et al. "Big self-supervised models are strong semi-supervised learners." (2020).

• Grill, Jean-Bastien, et al. "Bootstrap your own latent: A new approach to self-supervised learning." NeurIPS (2020).

11 – Variational Autoencoders (16/03/2022-21/03/2022)

In this lecture, we will discuss a family of latent variable models known as the Variational Autoencoders (VAE). We’ll see how a deep latent gaussian model can be seen as an autoencoder via amortized variational inference, and how such an autoencoder can be used as a generative model. At the end, we’ll take a look at variants of VAE and different ways to improve inference.

Slides:

• Variational Autoencoders

Video of lecture I of Variational Autoencoders (16/03/2022).
Video of lecture II of Variational Autoencoders (21/03/2022).

Reference: (* = you are responsible for this material)

• *Chapter 20.10.3 of the Deep Learning textbook.

• *Chapter 2 of An Introduction to Variational Autoencoders by Kingma and Welling

• Inference Suboptimality in Variational Autoencoders by Chris Cremer (ICML 2018)

• Importance Weighted Autoencoders by Yuri Burda (ICLR 2016)

• Variational Inference, lecture note by David Blei. Section 1-6.

• Blog post Variational Autoencoder Explained by Goker Erdogan

• Blog post Families of Generative Models by Andre Cianflone

09 – Autoencoders and Autoregressive Generative Models (07/03/2022-09/03/2022)

In this lecture we will take a closer look at a form of neural network known as an Autoencoder. We will also begin our look at generative models with Autoregressive Models.

Slides:

• Autoencoders

• Autoregressive Generative Models

Video of lecture 1 (07/03/2022).

Video of lecture 2 (09/03/2022).

Reference: (* = you are responsible for this material)

• *Chapter 13-14 of the Deep Learning textbook.

• *Sections 20.10.5-20.10.10 of the Deep Learning textbook.

• The Neural Autoregressive Distribution Estimator by Hugo Larochelle and Iain Murray (AISTAT2011) -- this is just a suggestion, we don't cover NADE in class.

• MADE: Masked Autoencoder for Distribution Estimation by Mathieu Germain, Karol Gregor, Iain Murray, Hugo Larochelle (ICML2015).

• Pixel Recurrent Neural Networks by Aaron van den Oord, Nal Kalchbrenner, Koray Kavukcuoglu (ICML2016)

• *Conditional Image Generation with PixelCNN Decoders by Aaron van den Oord, Nal Kalchbrenner, Oriol Vinyals, Lasse Espeholt, Alex Graves, Koray Kavukcuoglu (NIPS2016)

• *Medium Post by Jessica Dafflon on PixelCNN's blindspot and how to fix it.

07 – Self-Attention and Transformer (14/02/2022-16/02/2022)

In this talk Arian Hosseini will look at self-attention and the transformer model. We will see how they work, dig deep into them, see analysis and performances, and their applications mainly in natural language processing. We will see how some language models, based on transformer architecture, have surpassed human performance on some language understanding tasks, and we will also discuss their shortcomings.

Slides:

• Self-Attention and Transformer by A. Hosseini.

Video of lecture 1, part I (14/02/2022), chat (video ends abruptly due to technical issues)

Video of lecture 1, part 2 (14/02/2022)

Video of lecture 2 (16/02/2022)

Reference:

• Attention is all you need,

• BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding,

• Better Language Models and Their Implications,

• Improving Language Understanding by Generative Pre-Training,

05 – Sequential Models (02/02/2022, 07/02/2022, 09/02/2022)

In this lecture we introduce Recurrent Neural Networks and related models.

Lecture on RNNs (slides derived from Hugo Larochelle)

Video of lecture 1 (02/02/2022)
Video of lecture 2 (07/02/2022)
Video of lecture 3 (09/02/2022)

Reference: (* = you are responsible for this material)

• *Chapter 10 of the Deep Learning textbook (sections. 10.1-10.11, we will cover the material in 10.12 later).

• Blog post on Understanding LSTM Networks by Chris Olah.

03 – ConvNets (19/01/2022, 24/01/2022, 31/01/2022, 02/02/2022)

In this lecture we finish up our discussion of training neural networks and we introduce Convolutional Neural Networks.

Lecture 03 CNNs (some slides are modified from Hugo Larochelle’s course notes)

Backprop in CNNs (Slides are from Hiroshi Kuwajima’s Memo on Backpropagation in Convolutional Neural Networks.) -- I won't go over these in class, but they are required reading and you are responsible for it. (could be on assignments and exam)

Video of lecture 1 (19/01/2022)

Video of lecture 2 (24/01/2022)

Video of Lecture 3 (31/01/2022)

Video of Lecture 4 (02/02/2022)

Reference: (* = you are responsible for all of this material)

• *Chapter 9 of the Deep Learning textbook, Sections 9.10 and 9.11 are optional.

• Andrej Karpathy’s excellent tutorial on CNNs.

• Paper on convolution arithmetic by Vincent Dumoulin and Francesco Visin.

• WaveNet Blog presenting dilated convolutions animation and samples.

• Blog on Deconvolution and Checkerboard Artifacts by Augustus Odena, Vincent Dumoulin and Chris Olah.