A quantitative Robbins-Siegmund theorem
Neri, Morenikeji, Powell, Thomas
–arXiv.org Artificial Intelligence
The Robbins-Siegmund theorem is one of the most important results in stochastic optimization, where it is widely used to prove the convergence of stochastic algorithms. We provide a quantitative version of the theorem, establishing a bound on how far one needs to look in order to locate a region of metastability in the sense of Tao. Our proof involves a metastable analogue of Doob's theorem for $L_1$-supermartingales along with a series of technical lemmas that make precise how quantitative information propagates through sums and products of stochastic processes. In this way, our paper establishes a general methodology for finding metastable bounds for stochastic processes that can be reduced to supermartingales, and therefore for obtaining quantitative convergence information across a broad class of stochastic algorithms whose convergence proof relies on some variation of the Robbins-Siegmund theorem. We conclude by discussing how our general quantitative result might be used in practice.
arXiv.org Artificial Intelligence
Oct-21-2024
- Country:
- North America > United States
- New York (0.04)
- California (0.04)
- Michigan (0.04)
- Illinois (0.04)
- Europe
- United Kingdom > England
- Cambridgeshire > Cambridge (0.04)
- Germany > Hesse
- Darmstadt Region > Darmstadt (0.04)
- United Kingdom > England
- Asia
- Middle East > Israel (0.04)
- Japan > Honshū
- Kantō > Kanagawa Prefecture > Yokohama (0.04)
- North America > United States
- Genre:
- Research Report (0.50)