Darshil Doshi

According to our database, Darshil Doshi authored at least 6 papers between 2021 and 2024.

Collaborative distances:

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

2024

Grokking Modular Polynomials.

Darshil Doshi, Tianyu He, Aritra Das, Andrey Gromov

CoRR, 2024

Learning to grok: Emergence of in-context learning and skill composition in modular arithmetic tasks.

Tianyu He, Darshil Doshi, Aritra Das, Andrey Gromov

CoRR, 2024

To Grok or not to Grok: Disentangling Generalization and Memorization on Corrupted Algorithmic Datasets.

Darshil Doshi, Aritra Das, Tianyu He, Andrey Gromov

Proceedings of the Twelfth International Conference on Learning Representations, 2024

2023

Critical Initialization of Wide and Deep Neural Networks using Partial Jacobians: General Theory and Applications.

Darshil Doshi, Tianyu He, Andrey Gromov

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

2022

AutoInit: Automatic Initialization via Jacobian Tuning.

Tianyu He, Darshil Doshi, Andrey Gromov

CoRR, 2022

2021

Critical initialization of wide and deep neural networks through partial Jacobians: general theory and applications to LayerNorm.

Darshil Doshi, Tianyu He, Andrey Gromov

CoRR, 2021