Bildunterschrift
Figure 1: MVE training performance (top) and OVE training performance (bottom). We evaluate each for multiple rollout horizons H ∈ {1,3,5,10,20,30} and plot the mean and variance
across 5 random seeds.
Palenicek_Daniel_When to Trust Your Model in Model-based Value Expansion?
Palenicek_Daniel_When to Trust Your Model in Model-based Value Expansion?
