Deep Relaxation of Controlled Stochastic Gradient Descent via Singular Perturbations
Preprint
Abstract
We consider a singularly perturbed system of stochastic differential equations proposed by Chaudhari et al. (Res. Math. Sci. 2018) to approximate the Entropic Gradient Descent in the optimization of deep neural networks, via homogenisation. We embed it in a much larger class of two-scale stochastic control problems and rely on convergence results for Hamilton-Jacobi-Bellman equations with unbounded data proved recently by ourselves (ESAIM Control Optim. Calc. Var. 2023). We show that the limit of the value functions is itself the value function of an effective control problem with extended controls, and that the trajectories of the perturbed system converge in a suitable sense to the trajectories of the limiting effective control system. These rigorous results improve the understanding of the convergence of the algorithms used by Chaudhari et al., as well as of their possible extensions where some tuning parameters are modelled as dynamic controls.Keywords
All Related Versions
This publication has 18 references indexed in Scilit:
- On Local Entropy, Stochastic Control, and Deep Neural NetworksIEEE Control Systems Letters, 2022
- Entropic gradient descent algorithms and wide flat minima*Journal of Statistical Mechanics: Theory and Experiment, 2021
- Entropy-SGD: biasing gradient descent into wide valleys*Journal of Statistical Mechanics: Theory and Experiment, 2019
- Deep relaxation: partial differential equations for optimizing deep neural networksResearch in the Mathematical Sciences, 2018
- Subdominant Dense Clusters Allow for Simple Learning and High Computational Performance in Neural Networks with Discrete SynapsesPhysical Review Letters, 2015
- Multiscale Stochastic Volatility for Equity, Interest Rate, and Credit Derivatives2011
- Uniqueness of solutions of elliptic equations and uniqueness of invariant measures of diffusionsSbornik: Mathematics, 2002
- Introduction: Set-Valued Analysis in Control TheorySet-Valued Analysis, 2000
- Nonsmooth Analysis and Control Theory1998
- Weak Convergence Methods and Singularly Perturbed Stochastic Control and Filtering Problems1990