arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2312.03218cs.LGcs.CCstat.ML
收藏

采用自适应子空间搜索的加速梯度算法以实现实例更快优化

Accelerated Gradient Algorithms with Adaptive Subspace Search for Instance-Faster Optimization

Peking University(北京大学)

Yuanshi Liu, Hanzhen Zhao, Yang Xu, Pengyun Yue, Cong Fang

AI总结:

针对最坏情形复杂度无法反映具体问题难度的问题,本文引入 $(α, τ_α)$ 刻画 Hessian 退化并提出自适应子空间搜索加速梯度算法,在线性回归等任务上取得更优梯度复杂度。

摘要:

基于梯度的极小极大最优算法极大推动了连续优化与机器学习的发展。Yurii Nesterov 的一项开创性工作 [Nes83a] 为最小化一个 $L$-光滑、$μ$-强凸目标函数建立了 $\tilde{\mathcal{O}}(\sqrt{L/μ})$ 的梯度复杂度。然而,理想算法应当适应特定目标函数的显式复杂度,并在更简单问题上获得更快速率,这促使我们重新思考现有优化建模与分析的两点不足。(i)最坏情形最优性既不是实例最优性,也不是现实中的这种最优性。(ii)传统的 $L$-光滑性条件可能并非现代实际问题的首要抽象或刻画。本文中,我们开辟了一条设计和分析基于梯度算法的新途径,并可直接应用于包括线性回归及更广泛场景在内的机器学习。基于 Hessian 矩阵奇异值通常急剧下降这一观察,我们引入两个因子 $(α, τ_α)$,以更精细地描述优化问题的退化条件。我们设计了自适应算法,可在无需预先已知知识的情况下,以更少的梯度或类似 oracle 访问次数求解更简单的问题。这些算法还改进了机器学习中若干问题的当前最优复杂度,从而解决了在已知复杂度下界下如何设计更快算法这一开放问题。特别地,在 $\mathcal{O}(1)$-核范数有界条件下,我们在线性回归上实现了最优的 $\tilde{\mathcal{O}}(μ^{-1/3})$(相对于 $\tilde{\mathcal{O}}(μ^{-1/2})$)梯度复杂度。我们希望这项工作能够引发对优化中现代问题难度理解的重新思考。

英文摘要:

Gradient-based minimax optimal algorithms have greatly promoted the development of continuous optimization and machine learning. One seminal work due to Yurii Nesterov [Nes83a] established $\tilde{\mathcal{O}}(\sqrt{L/μ})$ gradient complexity for minimizing an $L$-smooth $μ$-strongly convex objective. However, an ideal algorithm would adapt to the explicit complexity of a particular objective function and incur faster rates for simpler problems, triggering our reconsideration of two defeats of existing optimization modeling and analysis. (i) The worst-case optimality is neither the instance optimality nor such one in reality. (ii) Traditional $L$-smoothness condition may not be the primary abstraction/characterization for modern practical problems. In this paper, we open up a new way to design and analyze gradient-based algorithms with direct applications in machine learning, including linear regression and beyond. We introduce two factors $(α, τ_α)$ to refine the description of the degenerated condition of the optimization problems based on the observation that the singular values of Hessian often drop sharply. We design adaptive algorithms that solve simpler problems without pre-known knowledge with reduced gradient or analogous oracle accesses. The algorithms also improve the state-of-art complexities for several problems in machine learning, thereby solving the open problem of how to design faster algorithms in light of the known complexity lower bounds. Specially, with the $\mathcal{O}(1)$-nuclear norm bounded, we achieve an optimal $\tilde{\mathcal{O}}(μ^{-1/3})$ (v.s. $\tilde{\mathcal{O}}(μ^{-1/2})$) gradient complexity for linear regression. We hope this work could invoke the rethinking for understanding the difficulty of modern problems in optimization.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑