A Near-optimal Algorithm for Learning Margin Halfspaces with Massart Noise

Authors: Ilias Diakonikolas, Nikos Zarifis

NeurIPS 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Theoretical Our main result is a computationally efficient learner with sample complexity eΘ(1/(γ2ϵ2)), nearly matching this lower bound. In addition, our algorithm is simple and practical, relying on online SGD on a carefully selected sequence of convex losses. The paper is theoretical in nature and does not include experiments.
Researcher Affiliation Academia Ilias Diakonikolas Department of Computer Sciences UW-Madison Madison, WI ilias@cs.wisc.edu Nikos Zarifis Department of Computer Sciences UW-Madison Madison, WI zarifis@wisc.edu
Pseudocode Yes Algorithm 1: Learning Margin Halfspaces with Massart Noise
Open Source Code No The paper is theoretical in nature and does not include experiments. The NeurIPS checklist responses within the paper indicate 'NA' for code and data access questions, and no explicit statement or link for code release is found.
Open Datasets No The paper is theoretical and analyzes algorithms in terms of sample complexity from a distribution D, but it does not specify or use any named public or open datasets for training.
Dataset Splits No The paper is theoretical and does not conduct experiments, therefore no training, validation, or test dataset splits are described.
Hardware Specification No The paper is theoretical in nature and does not include experiments, therefore no specific hardware specifications are provided.
Software Dependencies No The paper is theoretical and does not conduct experiments, therefore no specific software dependencies with version numbers are mentioned.
Experiment Setup No The paper is theoretical and does not describe an experimental setup with specific hyperparameters or training configurations.