Adapting to Mixing Time in Stochastic Optimization with Markovian Data

• URL: http://arxiv.org/abs/2202.04428v3
• Date: Thu, 13 Jul 2023 16:05:28 GMT
• Title: Adapting to Mixing Time in Stochastic Optimization with Markovian Data
• Authors: Ron Dorfman, Kfir Y. Levy
• Abstract summary: We consider optimization problems where data is drawn from a Markov chain. Existing methods for this setting crucially rely on knowing the mixing time of the chain. Our method relies on a novel combination of multi-level Monte Carlo (ML) gradient estimation together with an adaptive learning method.
• Score: 12.709177728330399
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: We consider stochastic optimization problems where data is drawn from a Markov chain. Existing methods for this setting crucially rely on knowing the mixing time of the chain, which in real-world applications is usually unknown. We propose the first optimization method that does not require the knowledge of the mixing time, yet obtains the optimal asymptotic convergence rate when applied to convex problems. We further show that our approach can be extended to: (i) finding stationary points in non-convex optimization with Markovian data, and (ii) obtaining better dependence on the mixing time in temporal difference (TD) learning; in both cases, our method is completely oblivious to the mixing time. Our method relies on a novel combination of multi-level Monte Carlo (MLMC) gradient estimation together with an adaptive learning method.

Related papers

• Effectively Leveraging Momentum Terms in Stochastic Line Search Frameworks for Fast Optimization of Finite-Sum Problems [0.5156484100374059]
  We explore the relationship between recent line search approaches for deep optimization in the overparametrized regime and momentum directions. We introduce algorithmic that exploits a mix of data persistency, conjugateient type rules for the definition of the momentum parameter. The resulting algorithm is empirically shown to outperform other popular methods.
  arXiv  Detail & Related papers  (2024-11-11T16:26:33Z)
• Fast Semisupervised Unmixing Using Nonconvex Optimization [80.11512905623417]
  We introduce a novel convex convex model for semi/library-based unmixing. We demonstrate the efficacy of Alternating Methods of sparse unsupervised unmixing.
  arXiv  Detail & Related papers  (2024-01-23T10:07:41Z)
• Stochastic-Constrained Stochastic Optimization with Markovian Data [2.1756081703276]
  We study the setting where data samples are drawn from a Markov chain and thus are not independent and identically distributed. We propose two variants of drift-plus-penalty; one is for the case when the mixing time of the underlying Markov chain is known. Our algorithms apply to a more general setting of constrained online convex optimization where the sequence of constraint functions follows a Markov chain.
  arXiv  Detail & Related papers  (2023-12-07T14:09:27Z)
• First Order Methods with Markovian Noise: from Acceleration to Variational Inequalities [91.46841922915418]
  We present a unified approach for the theoretical analysis of first-order variation methods. Our approach covers both non-linear gradient and strongly Monte Carlo problems. We provide bounds that match the oracle strongly in the case of convex method optimization problems.
  arXiv  Detail & Related papers  (2023-05-25T11:11:31Z)
• Beyond Exponentially Fast Mixing in Average-Reward Reinforcement Learning via Multi-Level Monte Carlo Actor-Critic [61.968469104271676]
  We propose an RL methodology attuned to the mixing time by employing a multi-level Monte Carlo estimator for the critic, the actor, and the average reward embedded within an actor-critic (AC) algorithm. We experimentally show that these alleviated restrictions on the technical conditions required for stability translate to superior performance in practice for RL problems with sparse rewards.
  arXiv  Detail & Related papers  (2023-01-28T04:12:56Z)
• The Curse of Unrolling: Rate of Differentiating Through Optimization [35.327233435055305]
  Un differentiation approximates the solution using an iterative solver and differentiates it through the computational path. We show that we can either 1) choose a large learning rate leading to a fast convergence but accept that the algorithm may have an arbitrarily long burn-in phase or 2) choose a smaller learning rate leading to an immediate but slower convergence.
  arXiv  Detail & Related papers  (2022-09-27T09:27:29Z)
• STORM+: Fully Adaptive SGD with Momentum for Nonconvex Optimization [74.1615979057429]
  We investigate non-batch optimization problems where the objective is an expectation over smooth loss functions. Our work builds on the STORM algorithm, in conjunction with a novel approach to adaptively set the learning rate and momentum parameters.
  arXiv  Detail & Related papers  (2021-11-01T15:43:36Z)
• Nonlinear Independent Component Analysis for Continuous-Time Signals [85.59763606620938]
  We study the classical problem of recovering a multidimensional source process from observations of mixtures of this process. We show that this recovery is possible for many popular models of processes (up to order and monotone scaling of their coordinates) if the mixture is given by a sufficiently differentiable, invertible function.
  arXiv  Detail & Related papers  (2021-02-04T20:28:44Z)
• Effective Proximal Methods for Non-convex Non-smooth Regularized Learning [27.775096437736973]
  We show that the independent sampling scheme tends to improve performance of the commonly-used uniform sampling scheme. Our new analysis also derives a speed for the sampling than best one available so far.
  arXiv  Detail & Related papers  (2020-09-14T16:41:32Z)
• Least Squares Regression with Markovian Data: Fundamental Limits and Algorithms [69.45237691598774]
  We study the problem of least squares linear regression where the data-points are dependent and are sampled from a Markov chain. We establish sharp information theoretic minimax lower bounds for this problem in terms of $tau_mathsfmix$. We propose an algorithm based on experience replay--a popular reinforcement learning technique--that achieves a significantly better error rate.
  arXiv  Detail & Related papers  (2020-06-16T04:26:50Z)

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.