Reinforcement Learning (RL) is a machine learning (ML) technique to learn sequential
decision-making in complex problems. RL is inspired by trial-and-error based human/animal …

StarCraft II is one of the most challenging simulated reinforcement learning environments; it
is partially observable, stochastic, multi-agent, and mastering StarCraft II requires strategic …

The Credit Assignment Problem (CAP) refers to the longstanding challenge of
Reinforcement Learning (RL) agents to associate actions with their long-term …

This paper proposes novel architectures for spatio-temporal graph convolutional and
recurrent neural networks whose structure is inspired by the physics of power systems. The …

StarCraft II is one of the most challenging reinforcement learning (RL) environments; it is
partially observable, stochastic, and multi-agent, and mastering StarCraft II requires strategic …

Off-policy sampling and experience replay are key for improving sample efficiency and
scaling model-free temporal difference learning methods. When combined with function …

Emphatic Temporal Diérence (TD) methods are a class of off-policy Reinforcement Learning
(RL) methods involving the use of followon traces. Despite the theoretical success of …

It is well known that Reinforcement Learning (RL) can be formulated as a convex program
with linear constraints. The dual form of this formulation is unconstrained, which we refer to …

We propose Heuristic Blending (HUBL), a simple performance-improving technique for a
broad class of offline RL algorithms based on value bootstrapping. HUBL modifies the …

Emphatic algorithms have shown great promise in stabilizing and improving reinforcement
learning by selectively emphasizing the update rule. Although the emphasis fundamentally …