Statistical Complexity and Optimal Algorithms for Non-linear Ridge Bandits

• URL: http://arxiv.org/abs/2302.06025v3
• Date: Wed, 10 Jan 2024 02:49:55 GMT
• Title: Statistical Complexity and Optimal Algorithms for Non-linear Ridge Bandits
• Authors: Nived Rajaraman, Yanjun Han, Jiantao Jiao, Kannan Ramchandran
• Abstract summary: We consider the sequential decision-making problem where the mean outcome is a non-linear function of the chosen action. In particular, a two-stage algorithm that first finds a good initial action and then treats the problem as locally linear is statistically optimal.
• Score: 39.391636494257284
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: We consider the sequential decision-making problem where the mean outcome is a non-linear function of the chosen action. Compared with the linear model, two curious phenomena arise in non-linear models: first, in addition to the "learning phase" with a standard parametric rate for estimation or regret, there is an "burn-in period" with a fixed cost determined by the non-linear function; second, achieving the smallest burn-in cost requires new exploration algorithms. For a special family of non-linear functions named ridge functions in the literature, we derive upper and lower bounds on the optimal burn-in cost, and in addition, on the entire learning trajectory during the burn-in period via differential equations. In particular, a two-stage algorithm that first finds a good initial action and then treats the problem as locally linear is statistically optimal. In contrast, several classical algorithms, such as UCB and algorithms relying on regression oracles, are provably suboptimal.

Related papers

• Pessimistic Nonlinear Least-Squares Value Iteration for Offline Reinforcement Learning [53.97335841137496]
  We propose an oracle-efficient algorithm, dubbed Pessimistic Least-Square Value Iteration (PNLSVI) for offline RL with non-linear function approximation. Our algorithm enjoys a regret bound that has a tight dependency on the function class complexity and achieves minimax optimal instance-dependent regret when specialized to linear function approximation.
  arXiv  Detail & Related papers  (2023-10-02T17:42:01Z)
• Online Learning Under A Separable Stochastic Approximation Framework [20.26530917721778]
  We propose an online learning algorithm for a class of machine learning models under a separable approximation framework. We show that the proposed algorithm produces more robust and test performance when compared to other popular learning algorithms.
  arXiv  Detail & Related papers  (2023-05-12T13:53:03Z)
• Accelerated First-Order Optimization under Nonlinear Constraints [73.2273449996098]
  We exploit between first-order algorithms for constrained optimization and non-smooth systems to design a new class of accelerated first-order algorithms. An important property of these algorithms is that constraints are expressed in terms of velocities instead of sparse variables.
  arXiv  Detail & Related papers  (2023-02-01T08:50:48Z)
• Nearly Minimax Optimal Reinforcement Learning for Linear Markov Decision Processes [80.89852729380425]
  We propose the first computationally efficient algorithm that achieves the nearly minimax optimal regret $tilde O(dsqrtH3K)$. Our work provides a complete answer to optimal RL with linear MDPs, and the developed algorithm and theoretical tools may be of independent interest.
  arXiv  Detail & Related papers  (2022-12-12T18:58:59Z)
• Implicit Parameter-free Online Learning with Truncated Linear Models [51.71216912089413]
  parameter-free algorithms are online learning algorithms that do not require setting learning rates. We propose new parameter-free algorithms that can take advantage of truncated linear models through a new update that has an "implicit" flavor. Based on a novel decomposition of the regret, the new update is efficient, requires only one gradient at each step, never overshoots the minimum of the truncated model, and retains the favorable parameter-free properties.
  arXiv  Detail & Related papers  (2022-03-19T13:39:49Z)
• Nearly Optimal Linear Convergence of Stochastic Primal-Dual Methods for Linear Programming [5.126924253766052]
  We show that the proposed method exhibits a linear convergence rate for solving sharp instances with a high probability. We also propose an efficient coordinate-based oracle for unconstrained bilinear problems.
  arXiv  Detail & Related papers  (2021-11-10T04:56:38Z)
• Single-Timescale Stochastic Nonconvex-Concave Optimization for Smooth Nonlinear TD Learning [145.54544979467872]
  We propose two single-timescale single-loop algorithms that require only one data point each step. Our results are expressed in a form of simultaneous primal and dual side convergence.
  arXiv  Detail & Related papers  (2020-08-23T20:36:49Z)
• A spectral algorithm for robust regression with subgaussian rates [0.0]
  We study a new linear up to quadratic time algorithm for linear regression in the absence of strong assumptions on the underlying distributions of samples. The goal is to design a procedure which attains the optimal sub-gaussian error bound even though the data have only finite moments.
  arXiv  Detail & Related papers  (2020-07-12T19:33:50Z)
• Resolving learning rates adaptively by locating Stochastic Non-Negative Associated Gradient Projection Points using line searches [0.0]
  Learning rates in neural network training are currently determined a priori to training using expensive manual or automated tuning. This study proposes gradient-only line searches to resolve the learning rate for neural network training algorithms.
  arXiv  Detail & Related papers  (2020-01-15T03:08:07Z)

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.