Online and Distribution-Free Robustness: Regression and Contextual Bandits with Huber Contamination

• URL: http://arxiv.org/abs/2010.04157v3
• Date: Thu, 10 Jun 2021 22:54:35 GMT
• Title: Online and Distribution-Free Robustness: Regression and Contextual Bandits with Huber Contamination
• Authors: Sitan Chen, Frederic Koehler, Ankur Moitra, Morris Yau
• Abstract summary: We revisit two classic high-dimensional online learning problems, namely linear regression and contextual bandits. We show that our algorithms succeed where conventional methods fail.
• Score: 29.85468294601847
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: In this work we revisit two classic high-dimensional online learning problems, namely linear regression and contextual bandits, from the perspective of adversarial robustness. Existing works in algorithmic robust statistics make strong distributional assumptions that ensure that the input data is evenly spread out or comes from a nice generative model. Is it possible to achieve strong robustness guarantees even without distributional assumptions altogether, where the sequence of tasks we are asked to solve is adaptively and adversarially chosen? We answer this question in the affirmative for both linear regression and contextual bandits. In fact our algorithms succeed where conventional methods fail. In particular we show strong lower bounds against Huber regression and more generally any convex M-estimator. Our approach is based on a novel alternating minimization scheme that interleaves ordinary least-squares with a simple convex program that finds the optimal reweighting of the distribution under a spectral constraint. Our results obtain essentially optimal dependence on the contamination level $\eta$, reach the optimal breakdown point, and naturally apply to infinite dimensional settings where the feature vectors are represented implicitly via a kernel map.

Related papers

• Distributionally Robust Skeleton Learning of Discrete Bayesian Networks [9.46389554092506]
  We consider the problem of learning the exact skeleton of general discrete Bayesian networks from potentially corrupted data. We propose to optimize the most adverse risk over a family of distributions within bounded Wasserstein distance or KL divergence to the empirical distribution. We present efficient algorithms and show the proposed methods are closely related to the standard regularized regression approach.
  arXiv  Detail & Related papers  (2023-11-10T15:33:19Z)
• Tight Certified Robustness via Min-Max Representations of ReLU Neural Networks [9.771011198361865]
  The reliable deployment of neural networks in control systems requires rigorous robustness guarantees. In this paper, we obtain tight robustness certificates over convex representations of ReLU neural networks.
  arXiv  Detail & Related papers  (2023-10-07T21:07:45Z)
• Provable Offline Preference-Based Reinforcement Learning [95.00042541409901]
  We investigate the problem of offline Preference-based Reinforcement Learning (PbRL) with human feedback. We consider the general reward setting where the reward can be defined over the whole trajectory. We introduce a new single-policy concentrability coefficient, which can be upper bounded by the per-trajectory concentrability.
  arXiv  Detail & Related papers  (2023-05-24T07:11:26Z)
• Implicitly normalized forecaster with clipping for linear and non-linear heavy-tailed multi-armed bandits [85.27420062094086]
  Implicitly Normalized Forecaster (INF) is considered an optimal solution for adversarial multi-armed bandit (MAB) problems. We propose a new version of INF called the Implicitly Normalized Forecaster with clipping (INFclip) for MAB problems with heavy-tailed settings. We demonstrate that INFclip is optimal for linear heavy-tailed MAB problems and works well for non-linear ones.
  arXiv  Detail & Related papers  (2023-05-11T12:00:43Z)
• Online Statistical Inference for Matrix Contextual Bandit [3.465827582464433]
  Contextual bandit has been widely used for sequential decision-making based on contextual information and historical feedback data. We introduce a new online doubly-debiasing inference procedure to simultaneously handle both sources of bias. Our inference results are built upon a newly developed low-rank gradient descent estimator and its non-asymptotic convergence result.
  arXiv  Detail & Related papers  (2022-12-21T22:03:06Z)
• On Kernelized Multi-Armed Bandits with Constraints [16.102401271318012]
  We study a bandit problem with a general unknown reward function and a general unknown constraint function. We propose a general framework for both algorithm performance analysis. We demonstrate the superior performance of our proposed algorithms via numerical experiments.
  arXiv  Detail & Related papers  (2022-03-29T14:02:03Z)
• Distribution-Free Robust Linear Regression [5.532477732693]
  We study random design linear regression with no assumptions on the distribution of the covariates. We construct a non-linear estimator achieving excess risk of order $d/n$ with the optimal sub-exponential tail. We prove an optimal version of the classical bound for the truncated least squares estimator due to Gy"orfi, Kohler, Krzyzak, and Walk.
  arXiv  Detail & Related papers  (2021-02-25T15:10:41Z)
• Nearly Dimension-Independent Sparse Linear Bandit over Small Action Spaces via Best Subset Selection [71.9765117768556]
  We consider the contextual bandit problem under the high dimensional linear model. This setting finds essential applications such as personalized recommendation, online advertisement, and personalized medicine. We propose doubly growing epochs and estimating the parameter using the best subset selection method.
  arXiv  Detail & Related papers  (2020-09-04T04:10:39Z)
• On Lower Bounds for Standard and Robust Gaussian Process Bandit Optimization [55.937424268654645]
  We consider algorithm-independent lower bounds for the problem of black-box optimization of functions having a bounded norm. We provide a novel proof technique for deriving lower bounds on the regret, with benefits including simplicity, versatility, and an improved dependence on the error probability.
  arXiv  Detail & Related papers  (2020-08-20T03:48:14Z)
• Approximation Schemes for ReLU Regression [80.33702497406632]
  We consider the fundamental problem of ReLU regression. The goal is to output the best fitting ReLU with respect to square loss given to draws from some unknown distribution.
  arXiv  Detail & Related papers  (2020-05-26T16:26:17Z)
• High-Dimensional Robust Mean Estimation via Gradient Descent [73.61354272612752]
  We show that the problem of robust mean estimation in the presence of a constant adversarial fraction can be solved by gradient descent. Our work establishes an intriguing connection between the near non-lemma estimation and robust statistics.
  arXiv  Detail & Related papers  (2020-05-04T10:48:04Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.