2 papers
cs.LG2026
Reflected Schrödinger Bridge Matching
Marcus Häggbom, Viktor Nilsson, Pierre Nyquist +1
Recent advances in generative modeling have enabled the efficient computation of Schrödinger bridges (SB) in high-dimensional settings by leveraging partially simulation-free train…
stat.ML2024
Mean-Field Microcanonical Gradient Descent
Marcus Häggbom, Morten Karlsmark, Joakim Andén
Microcanonical gradient descent is a sampling procedure for energy-based models allowing for efficient sampling of distributions in high dimension. It works by transporting samples…