SQ Lower Bounds for Learning Single Neurons with Massart Noise

Authors: Ilias Diakonikolas, Daniel Kane, Lisheng Ren, Yuxin Sun

NeurIPS 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Theoretical Our work is pure theoretical in nature. For a range of activation functions, including Re LUs, we establish superpolynomial Statistical Query (SQ) lower bounds for this learning problem.
Researcher Affiliation Academia Ilias Diakonikolas University of Wisconsin-Madison ilias@cs.wisc.edu Daniel M. Kane University of California, San Diego dakane@cs.ucsd.edu Lisheng Ren University of Wisconsin-Madison lren29@wisc.edu Yuxin Sun University of Wisconsin-Madison yxsun@cs.wisc.edu
Pseudocode No The paper describes its constructions and mathematical procedures using prose and definitions but does not include any formal pseudocode or algorithm blocks.
Open Source Code No The paper does not provide any statement about releasing source code for its methodology, nor does it provide a link to a code repository. The ethics checklist marks N/A for code.
Open Datasets No The paper is theoretical and focuses on lower bounds for learning problems with statistical queries; it does not use or refer to any publicly available or open datasets for empirical training or evaluation.
Dataset Splits No The paper is theoretical and does not perform experiments that require dataset splits for training, validation, or testing.
Hardware Specification No The paper is theoretical and does not describe any specific hardware used for experiments.
Software Dependencies No The paper is theoretical and does not mention any specific software dependencies or version numbers.
Experiment Setup No The paper is theoretical and does not describe an experimental setup with hyperparameters or system-level training settings.