Harsh Mehta

~Harsh_Mehta1

3

论文总数

3.0

年均投稿

6.1

平均评分

接收情况1/3

会议分布

ICLR

2

NeurIPS

1

发表论文 (3 篇)

20243 篇

When, Why and How Much? Adaptive Learning Rate Scheduling by Refinement

ICLR 2024Rejected

Convexifying Transformers: Improving optimization and understanding of transformer networks

ICLR 2024Rejected

The Road Less Scheduled

NeurIPS 2024Oral

合作者 (7)

Aaron Defazio2 篇

Ashok Cutkosky2 篇

Konstantin Mishchenko2 篇

Ahmed Khaled1 篇

Xingyu Alice Yang1 篇

Behnam Neyshabur1 篇

Tolga Ergen1 篇