SQ Lower Bounds for Non-Gaussian Component Analysis with Weaker Assumptions

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

NeurIPS 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Theoretical We study the complexity of Non-Gaussian Component Analysis (NGCA) in the Statistical Query (SQ) model. Prior work developed a general methodology to prove SQ lower bounds for this task... In this work, we establish that the latter condition is indeed not necessary. In particular, we prove near-optimal SQ lower bounds for NGCA under the moment-matching condition only.
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 consists of mathematical definitions, theorems, lemmas, and proofs. It does not include any sections labeled 'Pseudocode' or 'Algorithm', nor any structured algorithm blocks.
Open Source Code No The paper is theoretical and focuses on mathematical proofs of lower bounds. It does not describe an implementable software methodology and therefore does not provide any links or statements about releasing source code.
Open Datasets No The paper is theoretical and does not involve experimental evaluation with datasets. Consequently, there is no mention of dataset availability for training.
Dataset Splits No The paper is theoretical and does not involve empirical evaluation or dataset splits for validation. Therefore, no information on validation splits is provided.
Hardware Specification No The paper is purely theoretical and does not describe any computational experiments. Thus, there is no mention of hardware specifications used for running experiments.
Software Dependencies No The paper is theoretical and does not involve software implementation or computational experiments. Consequently, no software dependencies or their version numbers are mentioned.
Experiment Setup No The paper is theoretical and focuses on mathematical proofs and analyses. It does not describe any experimental setups, hyperparameters, or training configurations.