Year: 2015
Type: article
Abstract: We study accelerated mirror descent dynamics in continuous and discrete time. Combining the original continuous-time motivation of mirror descent with a recent ODE interpretation of Nesterov's accelerated method, we propose a family of continuous-time descent dynamics for convex functions with Lipschitz gradients, such that the solution trajectories converge to the optimum at a O(1/t2) rate. We then show that a large family of first-order accelerated methods can be obtained as a discretization of the ODE, and these methods converge at a O(1/k2) rate. This connection between accelerated mirror descent and the ODE provides an intuitive approach to the design and analysis of accelerated first-order algorithms.
Institution: University of California, Berkeley
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
Awards: Large-scale statistical machine learning, CPS: Frontiers: Collaborative Research: Foundations of Resilient CybEr-Physical Systems (FORCES), CPS: Frontiers: Collaborative Research: Foundations of Resilient CybEr-Physical Systems (FORCES), CPS: Frontiers: Collaborative Research: Foundations of Resilient CybEr-Physical Systems (FORCES), CPS: Frontiers: Collaborative Research: Foundations of Resilient CybEr-Physical Systems (FORCES), MCS: AF: Small: Algorithms for Large Scale Prediction Problems
OpenAlex ID: W2184596805
