Understanding the Impact of Model Incoherence on Convergence of Incremental SGD with Random Reshuffle
Jul 12, 2020
Speakers
Shaocong Ma
Speaker · 0 followers
About
Although SGD with random reshuffle is widely applied in machine learning applications, there is a limited understanding of how model characteristics affect the convergence of the algorithm. In this work, we introduce model incoherence to characterize the diversity of model characteristics and study its impact on convergence of SGD with random reshuffle. Specifically, minimizer incoherence measures the discrepancy between the global minimizer of a sample loss and that of the total loss and affects the convergence error of SGD. In particular, we show that the variable sequence generated by SGD with random reshuffle converges to a certain global minimizer of the total loss under full minimizer coherence. The other curvature incoherence measures the quality of condition numbers of the sample losses and determines the convergence rate of SGD. With model incoherence, our results show that SGD has a faster convergence rate and smaller convergence error under random reshuffle than those under random sampling, and hence provide theoretical justifications to the superior practical performance of SGD with random reshuffle. We further verify our theoretical results on model incoherence via experiments on training nonconvex machine learning models.Although SGD with random reshuffle is widely applied in machine learning applications, there is a limited understanding of how model characteristics affect the convergence of the algorithm. In this work, we introduce model incoherence to characterize the diversity of model characteristics and study its impact on convergence of SGD with random reshuffle. Specifically, minimizer incoherence measures the discrepancy between the global minimizer of a sample loss and that of the total loss and affect…
Categories
AI & Data Science
Category · 10.8k presentations
About ICML 2020
The International Conference on Machine Learning (ICML) is the premier gathering of professionals dedicated to the advancement of the branch of artificial intelligence known as machine learning. ICML is globally renowned for presenting and publishing cutting-edge research on all aspects of machine learning used in closely related areas like artificial intelligence, statistics and data science, as well as important application areas such as machine vision, computational biology, speech recognition, and robotics. ICML is one of the fastest growing artificial intelligence conferences in the world. Participants at ICML span a wide range of backgrounds, from academic and industrial researchers, to entrepreneurs and engineers, to graduate students and postdocs.
Like the format? Trust SlidesLive to capture your next event!
Professional recording and live streaming, delivered globally.
Sharing
Recommended Videos
Presentations on similar topic, category or speaker
Conservative Exploration in Bandits and Reinforcement Learning
Watch later
Learning Speech Representations from Raw Audio by Joint Audiovisual Self-Superision
Watch later
1min Intro
Watch later
