Survey examines trajectory density ratio estimation when data support is weak
A new arXiv survey reviews methods for comparing two stochastic processes via trajectory density ratios, a problem central to reinforcement learning and other scientific applications, with emphasis on settings with weak support.
Read the full story at arXiv stat.ML ↗
Timeline · 1 report
- 2026-11-10 05:00 · arXiv stat.ML
Trajectory Density Ratios under Weak Support: A Survey across the Sciences