Markov Chain Block Coordinate Descent
arXiv:1811.08990
Abstract
The method of block coordinate gradient descent (BCD) has been a powerful method for large-scale optimization. This paper considers the BCD method that successively updates a series of blocks selected according to a Markov chain. This kind of block selection is neither i.i.d. random nor cyclic. On the other hand, it is a natural choice for some applications in distributed optimization and Markov decision process, where i.i.d. random and cyclic selections are either infeasible or very expensive. By applying mixing-time properties of a Markov chain, we prove convergence of Markov chain BCD for minimizing Lipschitz differentiable functions, which can be nonconvex. When the functions are convex and strongly convex, we establish both sublinear and linear convergence rates, respectively. We also present a method of Markov chain inertial BCD. Finally, we discuss potential applications.
References in corpus (5)
- Even Faster Accelerated Coordinate Descent Using Non-Uniform Sampling
- Coordinate Descent Converges Faster with the Gauss-Southwell Rule Than Random Selection
- Asynchronous Coordinate Descent under More Realistic Assumptions
- Walkman: A Communication-Efficient Random-Walk Algorithm for Decentralized Optimization
- A2BCD: An Asynchronous Accelerated Block Coordinate Descent Algorithm With Optimal Complexity