Schedule
09:00 - 09:15
Introduction and opening remarks
09:15 - 09:45
Invited Talk 1 (Yingbin Liang)
Invited Talk 1 (Yingbin Liang)
09:45 - 10:00
Contributed Talk 1 : Universal Sharpness Dynamics in Neural Network Training: Fixed Point Analysis, Edge of Stability, and Route to Chaos
Contributed Talk 1 : Universal Sharpness Dynamics in Neural Network Training: Fixed Point Analysis, Edge of Stability, and Route to Chaos
10:00 - 11:00
Poster Session 1 and Coffee Break
Poster Session 1 and Coffee Break
11:00 - 11:15
Contributed Talk 2: Catapults in SGD: spikes in the training loss and their impact on generalization through feature learning
Contributed Talk 2: Catapults in SGD: spikes in the training loss and their impact on generalization through feature learning
11:15 - 11:30
Contributed Talk 3: Interpreting Grokked Transformers in Complex Modular Arithmetic
Contributed Talk 3: Interpreting Grokked Transformers in Complex Modular Arithmetic
11:30 - 11:45
Contributed Talk 4: What Can Transformer Learn with Varying Depth? Case Studies on Sequence Learning Tasks
Contributed Talk 4: What Can Transformer Learn with Varying Depth? Case Studies on Sequence Learning Tasks
11:45 - 12:00
Contributed Talk 5: Promises and Pitfalls of Generative Masked Language Modeling: Theoretical Framework and Practical Guidelines
Contributed Talk 5: Promises and Pitfalls of Generative Masked Language Modeling: Theoretical Framework and Practical Guidelines
12:00 - 14:00
Poster Session 2 and Lunch Break
14:00 - 14:30
Invited Talk 2 (Suvrit Sra)
Invited Talk 2 (Suvrit Sra)
14:30 - 15:00
Invited Talk 3 (Lester Mackey)
Invited Talk 3 (Lester Mackey)
15:00 - 15:30
Invited Talk 4 (Sanjeev Arora)
Invited Talk 4 (Sanjeev Arora)
15:30- 16:00
Poster Session 3 and Coffee Break
16:00 - 16:30
Invited Talk 5 (Mengdi Wang)
Invited Talk 5 (Mengdi Wang)
Contact the organizers: workshopbgpt@gmail.com