UMN Machine Learning Seminar
The UMN Machine Learning Seminar Series brings together faculty, students, and local industrial partners who are interested in the theoretical, computational, and applied aspects of machine learning, to pose problems, exchange ideas, and foster collaborations. The talks are every Thursday from 12 p.m. - 1 p.m. during the Summer 2021 semester.
This week's speaker, Rohan Anil (Google Brain) will be giving a talk titled "Scalable Second-Order Optimization for Deep Learning."
Optimization in machine learning, both theoretical and applied, is presently dominated by first-order gradient methods such as stochastic gradient descent. Second-order optimization methods, that involve second derivatives and/or second order statistics of the data, are far less prevalent despite strong theoretical properties, due to their prohibitive computation, memory and communication costs. In an attempt to bridge this gap between theoretical and practical optimization, we present a scalable implementation of a second-order preconditioned method (concretely, a variant of full-matrix Adagrad), that along with several critical algorithmic and numerical improvements, provides significant convergence and wall-clock time improvements compared to conventional first-order methods on state-of-the-art deep models. Our novel design effectively utilizes the prevalent heterogeneous hardware architecture for training deep models, consisting of a multicore CPU coupled with multiple accelerator units. We demonstrate superior performance compared to state-of-the-art on very large learning tasks such as machine translation with Transformers, language modeling with BERT, click-through rate prediction on Criteo, and image classification on ImageNet with ResNet-50.
Rohan Anil is a Senior Staff Software Engineer, Google Research, Brain Team. Lately, he has been working on scalable and practical optimization techniques for efficient training of neural networks in various regimes.