Feng Chen

According to our database, Feng Chen authored at least 5 papers between 2023 and 2025.

Collaborative distances:

• Dijkstra number of five.
• Erdős number of four.

2025

Rethinking Fine-Tuning when Scaling Test-Time Compute: Limiting Confidence Improves Mathematical Reasoning.

Feng Chen, Allan Raventós, Nan Cheng, Surya Ganguli, Shaul Druckmann

CoRR, February, 2025

2024

Get rich quick: exact solutions reveal how unbalanced initializations promote rapid feature learning.

Daniel Kunin, Allan Raventós, Clémentine C. J. Dominé, Feng Chen, David A. Klindt, Andrew M. Saxe, Surya Ganguli

Proceedings of the Advances in Neural Information Processing Systems 38: Annual Conference on Neural Information Processing Systems 2024, 2024

2023

Pretraining task diversity and the emergence of non-Bayesian in-context learning for regression.

Allan Raventós, Mansheej Paul, Feng Chen, Surya Ganguli

Proceedings of the Advances in Neural Information Processing Systems 36: Annual Conference on Neural Information Processing Systems 2023, 2023

Stochastic Collapse: How Gradient Noise Attracts SGD Dynamics Towards Simpler Subnetworks.

Feng Chen, Daniel Kunin, Atsushi Yamamura, Surya Ganguli

Proceedings of the Advances in Neural Information Processing Systems 36: Annual Conference on Neural Information Processing Systems 2023, 2023

Unmasking the Lottery Ticket Hypothesis: What's Encoded in a Winning Ticket's Mask?

Mansheej Paul, Feng Chen, Brett W. Larsen, Jonathan Frankle, Surya Ganguli, Gintare Karolina Dziugaite

Proceedings of the Eleventh International Conference on Learning Representations, 2023