Learning Stationary Markov Processes with Contrastive Adjustment
This work addresses the need for stationary Markov processes in human-computer design processes, enabling local exploration and iterative refinement, though it appears incremental as it builds on noise-annealed sampling.
The paper tackles the problem of learning Markov transition kernels that match the data distribution, introducing contrastive adjustment as a flexible optimization algorithm applicable to continuous and discrete state spaces, and demonstrates promising results on image synthesis tasks compared to state-of-the-art generative models.
We introduce a new optimization algorithm, termed contrastive adjustment, for learning Markov transition kernels whose stationary distribution matches the data distribution. Contrastive adjustment is not restricted to a particular family of transition distributions and can be used to model data in both continuous and discrete state spaces. Inspired by recent work on noise-annealed sampling, we propose a particular transition operator, the noise kernel, that can trade mixing speed for sample fidelity. We show that contrastive adjustment is highly valuable in human-computer design processes, as the stationarity of the learned Markov chain enables local exploration of the data manifold and makes it possible to iteratively refine outputs by human feedback. We compare the performance of noise kernels trained with contrastive adjustment to current state-of-the-art generative models and demonstrate promising results on a variety of image synthesis tasks.