Statistical applications of contrastive learning
This work provides a computationally feasible approach for statisticians and researchers dealing with complex models where traditional likelihood-based methods are impractical.
The paper addresses the computational intractability of likelihood functions in statistical models like energy-based and simulator-based models by applying contrastive learning as an alternative, demonstrating its use for parameter estimation, Bayesian inference, and experimental design.
The likelihood function plays a crucial role in statistical inference and experimental design. However, it is computationally intractable for several important classes of statistical models, including energy-based models and simulator-based models. Contrastive learning is an intuitive and computationally feasible alternative to likelihood-based learning. We here first provide an introduction to contrastive learning and then show how we can use it to derive methods for diverse statistical problems, namely parameter estimation for energy-based models, Bayesian inference for simulator-based models, as well as experimental design.