Autumn 2020 Discussions
We conduct two discussions every week where we dicuss the basic concepts and recent advancements in the field of Deep Learning.
Basic Discussions
We discuss a few fundamental concepts on Wednesdays.
| Date | Topic | Resources |
|---|---|---|
| 04-08-2020 | Linear Algebra | Slides |
| 11-08-2020 | Probability | Slides |
| 19-08-2020 | Neural Networks and CNNs | Slides |
| 26-08-2020 | Basic CNN architectures | Slides |
| 02-09-2020 | RNNs | Slides |
| 09-09-2020 | VAE | Slides |
| 16-09-2020 | GANs | Slides |
| 23-09-2020 | Embeddings | Slides |
| 30-09-2020 | Attention and Transformer | Slides |
| 07-10-2020 | ELMo and BERT | Slides |
| 14-10-2020 | RL-I: MDPs, Bellman Equations | Slides-A Slides-B |
| 11-11-2020 | Graph Neural Networks | Slides |
Advanced Discussions
We discuss the latest papers published in top tier conferences on Saturdays.
Date |
Paper 1 | Link |
Paper 2 | Link |
|---|---|---|---|---|
| 01-08-2020 | Reinforced active learning for image segmentation | https://arxiv.org/abs/2002.06583 | Enhanced POET: Open-Ended Reinforcement Learning through Unbounded Invention of Learning Challenges and their Solutions | https://arxiv.org/abs/2003.08536 |
| 08-08-2020 | Towards Recognizing Unseen Categories in Unseen Domains | https://arxiv.org/abs/2007.12256 | Neural Arithmetic Units | https://arxiv.org/abs/2001.05016 |
| 29-08-2020 | Learning Memory Access Patterns | https://arxiv.org/abs/1803.02329 | Equalization Loss for Long-Tailed Object Recognition | https://arxiv.org/abs/2003.05176 |
| 05-09-2020 | PnPNet: End-to-End Perception and Prediction with Tracking in the Loop | https://arxiv.org/abs/2005.14711 | CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Features | https://arxiv.org/abs/1905.048993 |
| 26-09-2020 | Adversarial Continual Learning | https://arxiv.org/abs/2003.09553 | Decentralized Reinforcement Learning: Global Decision-Making via Local Economic Transactions | https://arxiv.org/abs/2007.02382 |
| 10-10-2020 | Implicit Latent Variable Model for Scene-Consistent Motion Forecasting | https://arxiv.org/abs/2007.12036 | Large Batch Optimization for Deep Learning: Training BERT in 76 minutes | https://arxiv.org/abs/1904.00962 |
| 31-10-2020 | Causal Discovery with Reinforcement Learning | https://arxiv.org/abs/1906.04477 | What Should Not Be Contrastive in Contrastive Learning | https://arxiv.org/abs/2008.05659 |
| 07-11-2020 | Dual Super-Resolution Learning for Semantic Segmentation | https://openaccess.thecvf.com/content_CVPR_2020/papers/Wang_Dual_Super-Resolution_Learning_for_Semantic_Segmentation_CVPR_2020_paper.pdf | Neural Architecture Search without Training Learning | https://arxiv.org/abs/2006.04647 |
| 28-11-2020 | Super-Convergence: Very Fast Training of Neural Networks Using Large Learning Rates | https://arxiv.org/abs/1708.07120 | ELECTRA: Pre-training Text Encoders as Discriminators Rather Than Generators | https://iclr.cc/virtual_2020/poster_r1xMH1BtvB.html |
Vision and Language Group
Deep Learning Research Group