Skip to yearly menu bar Skip to main content


Poster
in
Workshop: Structured Probabilistic Inference and Generative Modeling

Fit Like You Sample: Sample-Efficient Generalized Score Matching from Fast Mixing Markov Chains

Yilong Qin · Andrej Risteski

Keywords: [ sample complexity ] [ score matching ] [ Theory ] [ annealing ]


Abstract: Score matching is an approach to learning probability distributions parametrized up to a constant of proportionality (e.g. EBMs). The idea is to fit the score of the distribution (i.e. xlogp(x)), rather than the likelihood, thus avoiding the need to evaluate the constant of proportionality. While there's a clear algorithmic benefit, the statistical "cost" can be steep: recent work by Koehler et al '23 showed that for distributions that have poor isoperimetric properties (a large Poincare or log-Sobolev constant), score matching is substantially statistically less efficient than maximum likelihood. However, many natural realistic distributions, e.g. multimodal distributions as simple as a mixture of two Gaussians---even in one dimension---have a poor Poincare constant.In this paper, we show a close connection between the mixing time of an arbitrary Markov process with generator L and a generalized score matching loss that tries to fit Opp. We instantiate this framework with several examples. In the special case of O=x, and L being the generator of Langevin diffusion, this generalizes and recovers the results from Koehler et al '23. If L corresponds to a Markov process corresponding to a continuous version of simulated tempering, we show the corresponding generalized score matching loss is a Gaussian-convolution annealed score matching loss, akin to the one proposed in Song-Ermon '19. Moreover, we show that if the distribution being learned is a mixture of K Gaussians in d dimensions, the sample complexity of annealed score matching is polynomial in d and K --- obviating the Poincar'e constant-based lower bounds of the basic score matching loss shown in Koehler et al. This is the first result characterizing the benefits of annealing for score matching---a crucial component in more sophisticated score-based approaches like Song-Ermon '19.

Chat is not available.