Emphatic Temporal-Difference Learning
It addresses the problem of instability in off-policy reinforcement learning for researchers, but is incremental as it primarily unifies and extends existing results.
The paper summarizes recent works on emphatic temporal-difference learning algorithms, which achieve stability and convergence under off-policy training with linear function approximation, and demonstrates empirical benefits such as state-dependent discounting and resource allocation.
Emphatic algorithms are temporal-difference learning algorithms that change their effective state distribution by selectively emphasizing and de-emphasizing their updates on different time steps. Recent works by Sutton, Mahmood and White (2015), and Yu (2015) show that by varying the emphasis in a particular way, these algorithms become stable and convergent under off-policy training with linear function approximation. This paper serves as a unified summary of the available results from both works. In addition, we demonstrate the empirical benefits from the flexibility of emphatic algorithms, including state-dependent discounting, state-dependent bootstrapping, and the user-specified allocation of function approximation resources.