Statistical physics of inference: Thresholds and algorithms

L Zdeborová, F Krzakala - Advances in Physics, 2016 - Taylor & Francis
Many questions of fundamental interest in today's science can be formulated as inference
problems: some partial, or noisy, observations are performed over a set of variables and the …

Planting undetectable backdoors in machine learning models

S Goldwasser, MP Kim… - 2022 IEEE 63rd …, 2022 - ieeexplore.ieee.org
Given the computational cost and technical expertise required to train machine learning
models, users may delegate the task of learning to a service provider. Delegation of learning …

Unsupervised alignment of embeddings with wasserstein procrustes

E Grave, A Joulin, Q Berthet - The 22nd International …, 2019 - proceedings.mlr.press
We consider the task of aligning two sets of points in high dimension, which has many
applications in natural language processing and computer vision. As an example, it was …

Fast low-rank estimation by projected gradient descent: General statistical and algorithmic guarantees

Y Chen, MJ Wainwright - arxiv preprint arxiv:1509.03025, 2015 - arxiv.org
Optimization problems with rank constraints arise in many applications, including matrix
regression, structured PCA, matrix completion and matrix decomposition problems. An …

Statistical query lower bounds for robust estimation of high-dimensional gaussians and gaussian mixtures

I Diakonikolas, DM Kane… - 2017 IEEE 58th Annual …, 2017 - ieeexplore.ieee.org
We describe a general technique that yields the first Statistical Query lower bounds for a
range of fundamental high-dimensional learning problems involving Gaussian distributions …

The computational complexity of the restricted isometry property, the nullspace property, and related concepts in compressed sensing

AM Tillmann, ME Pfetsch - IEEE Transactions on Information …, 2013 - ieeexplore.ieee.org
This paper deals with the computational complexity of conditions which guarantee that the
NP-hard problem of finding the sparsest solution to an underdetermined linear system can …

Tensor SVD: Statistical and computational limits

A Zhang, D **a - IEEE Transactions on Information Theory, 2018 - ieeexplore.ieee.org
In this paper, we propose a general framework for tensor singular value decomposition
(tensor singular value decomposition (SVD)), which focuses on the methodology and theory …

Notes on computational hardness of hypothesis testing: Predictions using the low-degree likelihood ratio

D Kunisky, AS Wein, AS Bandeira - ISAAC Congress (International Society …, 2019 - Springer
These notes survey and explore an emerging method, which we call the low-degree
method, for understanding statistical-versus-computational tradeoffs in high-dimensional …