Machine Learning Department
School of Computer Science, Carnegie Mellon University


New Optimization Methods for Modern Machine Learning

Sashank J. Reddi

July 2017

Ph.D. Thesis


Keywords: Machine Learning, Optimization, Large-scale, Distributed optimization, Communication-efficient, Finite-sum, Variance-reduction, Bayesian inference

Modern machine learning systems pose several new statistical, scalability, privacy and ethical challenges. With the advent of massive datasets and increasingly complex tasks, scalability has especially become a critical issue in these systems. In this thesis, we focus on fundamental challenges related to scalability, such as computational and communication efficiency, in modern machine learning applications. The underlying central message of this thesis is that classical statistical thinking leads to highly effective optimization methods for modern big data applications.

The first part of the thesis investigates optimization methods for solving large-scale nonconvex Empirical Risk Minimization (ERM) problems. Such problems have surged into prominence, notably through deep learning, and have led to exciting progress. However, our understanding of optimization methods suitable for these problems is still very limited. We develop and analyze a new line of optimization methods for nonconvex ERM problems, based on the principle of variance reduction. We show that our methods exhibit fast convergence to stationary points and improve the state-of-the-art in several nonconvex ERM settings, including nonsmooth and constrained ERM. Using similar principles, we also develop novel optimization methods that provably converge to second-order stationary points. Finally, we show that the key principles behind our methods can be generalized to overcome challenges in other important problems such as Bayesian inference.

The second part of the thesis studies two critical aspects of modern distributed machine learning systems–asynchronicity and communication efficiency of optimization methods. We study various asynchronous stochastic algorithms with fast convergence for convex ERM problems and show that these methods achieve near-linear speedups in sparse settings common to machine learning. Another key factor governing the overall performance of a distributed system is its communication efficiency. Traditional optimization algorithms used in machine learning are often ill-suited for distributed environments with high communication cost. To address this issue, we discuss two different paradigms to achieve communication efficiency of algorithms in distributed environments and explore new algorithms with better communication complexity.

282 pages

Thesis Committee:
Alexander J. Smola (Co-Chair)
Barnabá P&oacoute;czos (Co-Chair)
Geoffrey J. Gordon
Suvrit Sra (Massachusetts Institute of Technology)
Stephen Boyd (Stanford University)

Manuela M. Veloso, Head, Machine Learning Department
Andrew W. Moore, Dean, School of Computer Science

SCS Technical Report Collection
School of Computer Science