Generative modeling with projected entangled-pair states
arXiv:2202.08177v125 citations
Originality Incremental advance
AI Analysis
This addresses the problem of improving generative modeling for image data, but it appears incremental as it builds on existing PEPS sampling methods.
The paper tackled generative modeling of 2D-structured datasets like images by using projected entangled-pair states (PEPS), demonstrating that PEPS significantly outperform matrix product states in this task.
We argue and demonstrate that projected entangled-pair states (PEPS) outperform matrix product states significantly for the task of generative modeling of datasets with an intrinsic two-dimensional structure such as images. Our approach builds on a recently introduced algorithm for sampling PEPS, which allows for the efficient optimization and sampling of the distributions.