Therefore, implicit bias in non-smooth neural networks trained by gradient descent remains an open question.

The deviated model can be motivated by many applications in science.

Sparse linear regression is a central problem in high-dimensional statistics. We study the correlated random design setting, where the covariates are drawn from a multivariate Gaussian N (0, Σ), and we seek an estimator with small excess risk. If the true signal is t -sparse, information-theoretically, it is possible to achieve strong recovery guarantees with only O (t log n) samples. However, computationally efficient algorithms have sample complexity linear in (some variant of) the condition number of Σ . Classical algorithms such as the Lasso can require significantly more samples than necessary even if there is only a single sparse approximate dependency among the covariates. We provide a polynomial-time algorithm that, given Σ, automatically adapts the Lasso to tolerate a small number of approximate dependencies. In particular, we achieve near-optimal sample complexity for constant sparsity and if Σ has few "outlier" eigenvalues. Our algorithm fits into a broader framework of feature adaptation for sparse linear regression with ill-conditioned covariates. With this framework, we additionally provide the first polynomial-factor improvement over brute-force search for constant sparsity t and arbitrary covariance Σ .

Total size 5.29GB Physiological total duration across subjects and across tasks 1119 min Video total duration across subjects and across tasks 918 min Audio total duration across subjects and across tasks 385 minFigure 1: A dataset summary card for StressID, constructed based on [2, 5]. 3 Figure 2: Organisation of the

Additionally, we provide multimodal predictive models combining video, audio, and physiological inputs.