1 paper · 2 filters
Humphrey Munn, Brendan Tidd, Peter Bohm +2
Deploying learned control policies is risky because policies that appear robust in simulation can confidently enter out-of-distribution (OOD) states after Sim-to-Real transfer, cau…