| layout |
|---|
default |
The recent two decades have seen a growing interest in the explicit modeling of causal structure in the application and development of novel inference techniques. At the same time, fueled by the success of deep learning, purely data-driven inference approaches have seen wide-spread adaptation in numerous domains. While these two developments happened largely in parallel, the respective communities have worked largely detached from each other. Only recently, there has been an increased interest addressing research problems at the intersection of causal inference and deep learning.
- Instructor: Igor Gilitschenski (gilitschenski@cs.toronto.edu, Office Hourse: TBD)
- TA: Ashkan Mirzaei (a.mirzaei@mail.utoronto.edu, Office Hourse: TBD)
This course will introduce the basic terminology from Causal Inference underlying these works and subsequently discuss some of the recent approaches that combine ideas from both communities in a seminar-style setting. The offered topics can involve Causal Inference-informed approaches in Domain Adaptation, Deep Reinforcement and Imitation Learning, Causal Discovery, Counterfactual Data Augmentation, AI Fairness, Adversarial Robustness, and Causal Inference for Computer Vision and Perception.
| Week | Date | Paper |
|---|---|---|
| 1 | Jan 12 | TBD |
| 2 | Jan 19 | TBD |
| 3 | Jan 26 | TBD |