In general, research in distributional reinforcement learning has focused on the classical setting of a scalar reward function.

In such situations, the successor representation (SR) is a popular framework which supports rapid policy evaluation by decoupling a policy's

Delusionalbias ariseswhenthe approximation architecture limits the class ofexpressible greedy policies.

Additionally, our approach accelerates learning of walking tasks and increases learning success rate.

Most existing work in this area relies on non-probabilistic approaches or on complex Reinforcement Learning (RL)-based hard structures.

Rashomon sets can be large in size and complicated in structure, particularly for highly nonlinear function classes that allow complexinteraction terms, such asdecision trees.