Dravyansh Sharma

Postdoc@Toyota Technological Institute at Chicago·美国·OpenReview

研究方向

learning theory · adversarial machine learning · semi-supervised learning · active learning · optimization · online learning · speech · natural language processing

7.8

Conservative classifiers do consistently well with improving agents: characterizing statistical and online learning

NeurIPS 2025Spotlight

一作

6.8

Sample complexity of data-driven tuning of model hyperparameters in neural networks with structured parameter-dependent dual function

NeurIPS 2025Poster

三作

6.4

Theoretical Analyses of Hyperparameter Selection in Graph-Based Semi-Supervised Learning

ICLR 2025Rejected

三作

6.0

Accelerating ERM for data-driven algorithm design using output-sensitive techniques

合作者 (12)

Dravyansh Sharma

Conservative classifiers do consistently well with improving agents: characterizing statistical and online learning

Sample complexity of data-driven tuning of model hyperparameters in neural networks with structured parameter-dependent dual function

On Learning Verifiers and Implications to Chain-of-Thought Reasoning

PAC Learning with Improvements

Provable Data-driven Hyperparameter Tuning for Deep Neural Networks

Theoretical Analyses of Hyperparameter Selection in Graph-Based Semi-Supervised Learning

Accelerating ERM for data-driven algorithm design using output-sensitive techniques