Efficient Stochastic Optimisation via Sequential Monte Carlo
This addresses computational bottlenecks in stochastic optimization for machine learning and statistics, though it appears incremental as it adapts existing SMC methods to a specific problem.
The paper tackles the problem of optimizing functions with intractable gradients, such as in maximum likelihood estimation and generative model fine-tuning, by developing sequential Monte Carlo (SMC) samplers to replace expensive inner sampling loops, resulting in significant computational gains.
The problem of optimising functions with intractable gradients frequently arise in machine learning and statistics, ranging from maximum marginal likelihood estimation procedures to fine-tuning of generative models. Stochastic approximation methods for this class of problems typically require inner sampling loops to obtain (biased) stochastic gradient estimates, which rapidly becomes computationally expensive. In this work, we develop sequential Monte Carlo (SMC) samplers for optimisation of functions with intractable gradients. Our approach replaces expensive inner sampling methods with efficient SMC approximations, which can result in significant computational gains. We establish convergence results for the basic recursions defined by our methodology which SMC samplers approximate. We demonstrate the effectiveness of our approach on the reward-tuning of energy-based models within various settings.