Learning Latent Energy-Based Models via Interacting Particle Langevin Dynamics
This work addresses the problem of efficient learning in latent energy-based models for researchers in machine learning, representing an incremental improvement through a novel algorithmic framework.
The paper tackled learning latent variable models with energy-based priors by developing interacting particle algorithms based on stochastic differential equations, resulting in a practical method with theoretical convergence guarantees and empirical validation on synthetic and image datasets.
We develop interacting particle algorithms for learning latent variable models with energy-based priors. To do so, we leverage recent developments in particle-based methods for solving maximum marginal likelihood estimation (MMLE) problems. Specifically, we provide a continuous-time framework for learning latent energy-based models, by defining stochastic differential equations (SDEs) that provably solve the MMLE problem. We obtain a practical algorithm as a discretisation of these SDEs and provide theoretical guarantees for the convergence of the proposed algorithm. Finally, we demonstrate the empirical effectiveness of our method on synthetic and image datasets.