04:00
2026-08-05
machinebrief.com
machine-learning
Convex-Hull-Neighborhood Smooth Dual Generalization: Controlling Local Correction Propagation in Offline RL
Researchers propose Convex Hull Neighborhood Smooth Dual Generalization (CSDG), a new offline reinforcement learning method that explicitly separates in-sample value targets from local out-of-distribu…