ResearchTrend.AI
  • Papers
  • Communities
  • Events
  • Blog
  • Pricing
Papers
Communities
Social Events
Terms and Conditions
Pricing
Parameter LabParameter LabTwitterGitHubLinkedInBlueskyYoutube

© 2025 ResearchTrend.AI, All rights reserved.

  1. Home
  2. Papers
  3. 2207.08540
27
16

Multi-block-Single-probe Variance Reduced Estimator for Coupled Compositional Optimization

18 July 2022
Wei Jiang
Gang Li
Yibo Wang
Lijun Zhang
Tianbao Yang
ArXivPDFHTML
Abstract

Variance reduction techniques such as SPIDER/SARAH/STORM have been extensively studied to improve the convergence rates of stochastic non-convex optimization, which usually maintain and update a sequence of estimators for a single function across iterations. What if we need to track multiple functional mappings across iterations but only with access to stochastic samples of O(1)\mathcal{O}(1)O(1) functional mappings at each iteration? There is an important application in solving an emerging family of coupled compositional optimization problems in the form of ∑i=1mfi(gi(w))\sum_{i=1}^m f_i(g_i(\mathbf{w}))∑i=1m​fi​(gi​(w)), where gig_igi​ is accessible through a stochastic oracle. The key issue is to track and estimate a sequence of g(w)=(g1(w),…,gm(w))\mathbf g(\mathbf{w})=(g_1(\mathbf{w}), \ldots, g_m(\mathbf{w}))g(w)=(g1​(w),…,gm​(w)) across iterations, where g(w)\mathbf g(\mathbf{w})g(w) has mmm blocks and it is only allowed to probe O(1)\mathcal{O}(1)O(1) blocks to attain their stochastic values and Jacobians. To improve the complexity for solving these problems, we propose a novel stochastic method named Multi-block-Single-probe Variance Reduced (MSVR) estimator to track the sequence of g(w)\mathbf g(\mathbf{w})g(w). It is inspired by STORM but introduces a customized error correction term to alleviate the noise not only in stochastic samples for the selected blocks but also in those blocks that are not sampled. With the help of the MSVR estimator, we develop several algorithms for solving the aforementioned compositional problems with improved complexities across a spectrum of settings with non-convex/convex/strongly convex/Polyak-{\L}ojasiewicz (PL) objectives. Our results improve upon prior ones in several aspects, including the order of sample complexities and dependence on the strong convexity parameter. Empirical studies on multi-task deep AUC maximization demonstrate the better performance of using the new estimator.

View on arXiv
Comments on this paper