SCIS - Schedule

Saturday, July 29th, 2023 (all times are in HST, GMT-10)

8.50-9.00am - Introductory remarks

9.00-9.45am - Francesco Locatello
Distribution Shifts in Generalist and Causal Models

10.15-10.30am - Break

10.30-11.15am - Sanmi Koyejo
On learning domain general predictors

11.15-12.00pm - Suchi Saria and Adarsh Subbaswamy
Using Causality to Improve Safety Throughout the AI Lifecycle

12.00-1.00pm - Lunch Break

1.00-1.45pm - Ludwig Schmidt
A data-centric view on reliable generalization: From ImageNet to LAION-5B

1.45-2.30pm - Sara Magliacane
Causal vs Causality-inspired representation learning

2.30-3.30pm - Poster Session 1

3.30-4.15pm - SCIS 2023 Panel, The Future of Generalization: Scale, Safety and Beyond

Adam Gleave, Moderator

Participants:

Samuel Bowman
Maggie Makar
Zachary Lipton

4.15-5.00pm - Casual Conversation + Poster Session 2

Saturday, July 29th, 2023 (all times are in HST, GMT-10)

8.50-9.00am - Introductory remarks

9.00-9.45am - Francesco Locatello
Distribution Shifts in Generalist and Causal Models

Provable domain adaptation using privileged information
Adam Breitholtz, Anton Matsson, Fredrik D. Johansson

Delphic Offline Reinforcement Learning under Nonidentifiable Hidden Confounding
Alizée Pace, Hugo Yèche, Bernhard Schölkopf, Gunnar Ratsch, Guy Tennenholtz

ModelDiff: A Framework for Comparing Learning Algorithms
Harshay Shah, Sung Min Park, Andrew Ilyas, Aleksander Madry

10.15-10.30am - Break

10.30-11.15am - Sanmi Koyejo
On learning domain general predictors

10.30-11.15am - Sanmi Koyejo
On learning domain general predictors

11.15-12.00pm - Suchi Saria and Adarsh Subbaswamy
Using Causality to Improve Safety Throughout the AI Lifecycle

12.00-1.00pm - Lunch Break

1.00-1.45pm - Ludwig Schmidt
A data-centric view on reliable generalization: From ImageNet to LAION-5B

1.45-2.30pm - Sara Magliacane
Causal vs Causality-inspired representation learning

2.30-3.30pm - Poster Session 1

3.30-4.15pm - SCIS 2023 Panel, The Future of Generalization: Scale, Safety and Beyond

Adam Gleave, Moderator

4.15-5.00pm - Casual Conversation + Poster Session 2

Saturday, July 29th, 2023 (all times are in HST, GMT-10)

8.50-9.00am - Introductory remarks

9.00-9.45am - Francesco Locatello Distribution Shifts in Generalist and Causal Models

Provable domain adaptation using privileged informationAdam Breitholtz, Anton Matsson, Fredrik D. Johansson

Delphic Offline Reinforcement Learning under Nonidentifiable Hidden Confounding Alizée Pace, Hugo Yèche, Bernhard Schölkopf, Gunnar Ratsch, Guy Tennenholtz ModelDiff: A Framework for Comparing Learning Algorithms Harshay Shah, Sung Min Park, Andrew Ilyas, Aleksander Madry

10.15-10.30am - Break

10.30-11.15am - Sanmi Koyejo On learning domain general predictors

10.30-11.15am - Sanmi Koyejo On learning domain general predictors

11.15-12.00pm - Suchi Saria and Adarsh Subbaswamy Using Causality to Improve Safety Throughout the AI Lifecycle

12.00-1.00pm - Lunch Break

1.00-1.45pm - Ludwig Schmidt A data-centric view on reliable generalization: From ImageNet to LAION-5B

1.45-2.30pm - Sara Magliacane Causal vs Causality-inspired representation learning

2.30-3.30pm - Poster Session 1

3.30-4.15pm - SCIS 2023 Panel, The Future of Generalization: Scale, Safety and BeyondAdam Gleave, Moderator

4.15-5.00pm - Casual Conversation + Poster Session 2

9.00-9.45am - Francesco Locatello
Distribution Shifts in Generalist and Causal Models

Provable domain adaptation using privileged information
Adam Breitholtz, Anton Matsson, Fredrik D. Johansson

Delphic Offline Reinforcement Learning under Nonidentifiable Hidden Confounding
Alizée Pace, Hugo Yèche, Bernhard Schölkopf, Gunnar Ratsch, Guy Tennenholtz

ModelDiff: A Framework for Comparing Learning Algorithms
Harshay Shah, Sung Min Park, Andrew Ilyas, Aleksander Madry

10.30-11.15am - Sanmi Koyejo
On learning domain general predictors

10.30-11.15am - Sanmi Koyejo
On learning domain general predictors

11.15-12.00pm - Suchi Saria and Adarsh Subbaswamy
Using Causality to Improve Safety Throughout the AI Lifecycle

1.00-1.45pm - Ludwig Schmidt
A data-centric view on reliable generalization: From ImageNet to LAION-5B

1.45-2.30pm - Sara Magliacane
Causal vs Causality-inspired representation learning

3.30-4.15pm - SCIS 2023 Panel, The Future of Generalization: Scale, Safety and Beyond

Adam Gleave, Moderator