The All-or-Nothing Phenomenon in Sparse Tensor PCA

Authors: Jonathan Niles-Weed, Ilias Zadik

NeurIPS 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Theoretical We study the statistical problem of estimating a rank-one sparse tensor corrupted by additive gaussian noise, a Gaussian additive model also known as sparse tensor PCA. We show that for Bernoulli and Bernoulli-Rademacher distributed signals and for all sparsity levels which are sublinear in the dimension of the signal, the sparse tensor PCA model exhibits a phase transition called the all-or-nothing phenomenon. This is the property that for some signal-to-noise ratio (SNR) SNRc and any fixed > 0, if the SNR of the model is below (1 ) SNRc, then it is impossible to achieve any arbitrarily small constant correlation with the hidden signal, while if the SNR is above (1 + ) SNRc, then it is possible to achieve almost perfect correlation with the hidden signal. Our results follow from a more general result showing that for any Gaussian additive model with a discrete uniform prior, the all-or-nothing phenomenon follows as a direct outcome of an appropriately defined near-orthogonality property of the support of the prior distribution.
Researcher Affiliation Academia Jonathan Niles-Weed Courant Insititute of Mathematical Scienes and Center for Data Science New York University jnw@cims.nyu.edu Ilias Zadik Center for Data Science New York University zadik@nyu.edu
Pseudocode No The paper does not contain structured pseudocode or algorithm blocks. It is a theoretical paper focusing on mathematical proofs.
Open Source Code No The paper does not provide concrete access to source code for the methodology described.
Open Datasets No The paper does not use or provide concrete access information for a publicly available or open dataset. It deals with theoretical models and priors (Bernoulli, Bernoulli-Rademacher) rather than empirical datasets.
Dataset Splits No The paper does not describe dataset splits for reproduction as it does not conduct experiments on empirical datasets.
Hardware Specification No The paper does not provide specific hardware details as it does not report on computational experiments requiring such specifications.
Software Dependencies No The paper does not provide specific ancillary software details with version numbers, as it is a theoretical work and does not report on computational experiments requiring such reproducibility.
Experiment Setup No The paper does not contain specific experimental setup details, such as hyperparameter values or training configurations, as it is a theoretical work and does not report on empirical experiments.