A framework for bilevel optimization that enables stochastic and global variance reduction algorithms

by   Mathieu Dagréou, et al.

Bilevel optimization, the problem of minimizing a value function which involves the arg-minimum of another function, appears in many areas of machine learning. In a large scale setting where the number of samples is huge, it is crucial to develop stochastic methods, which only use a few samples at a time to progress. However, computing the gradient of the value function involves solving a linear system, which makes it difficult to derive unbiased stochastic estimates. To overcome this problem we introduce a novel framework, in which the solution of the inner problem, the solution of the linear system, and the main variable evolve at the same time. These directions are written as a sum, making it straightforward to derive unbiased estimates. The simplicity of our approach allows us to develop global variance reduction algorithms, where the dynamics of all variables is subject to variance reduction. We demonstrate that SABA, an adaptation of the celebrated SAGA algorithm in our framework, has O(1/T) convergence rate, and that it achieves linear convergence under Polyak-Lojasciewicz assumption. This is the first stochastic algorithm for bilevel optimization that verifies either of these properties. Numerical experiments validate the usefulness of our method.


page 1

page 2

page 3

page 4


Stochastic Variance Reduction Methods for Policy Evaluation

Policy evaluation is a crucial step in many reinforcement-learning proce...

Unbiased Simulation for Optimizing Stochastic Function Compositions

In this paper, we introduce an unbiased gradient simulation algorithms f...

Improving KernelSHAP: Practical Shapley Value Estimation via Linear Regression

The Shapley value solution concept from cooperative game theory has beco...

Distributed stochastic gradient tracking algorithm with variance reduction for non-convex optimization

This paper proposes a distributed stochastic algorithm with variance red...

SVRG for Policy Evaluation with Fewer Gradient Evaluations

Stochastic variance-reduced gradient (SVRG) is an optimization method or...

SUCAG: Stochastic Unbiased Curvature-aided Gradient Method for Distributed Optimization

We propose and analyze a new stochastic gradient method, which we call S...

A Block Coordinate Ascent Algorithm for Mean-Variance Optimization

Risk management in dynamic decision problems is a primary concern in man...

Please sign up or login with your details

Forgot password? Click here to reset