AICurious Logo

What is: Replacing Eligibility Trace?

Year2000
Data SourceCC BY-SA - https://paperswithcode.com

In a Replacing Eligibility Trace, each time the state is revisited, the trace is reset to 11 regardless of the presence of a prior trace.. For the memory vector e_tRb0\textbf{e}\_{t} \in \mathbb{R}^{b} \geq \textbf{0}:

e_0=0\mathbf{e\_{0}} = \textbf{0}

e_t=γλe_t1(s) if ss_t\textbf{e}\_{t} = \gamma\lambda{e}\_{t-1}\left(s\right) \text{ if } s \neq s\_{t}

e_t=1 if s=s_t\textbf{e}\_{t} = 1 \text{ if } s = s\_{t}

They can be seen as crude approximation to dutch traces, which have largely superseded them as they perform better than replacing traces and have a clearer theoretical basis. Accumulating traces remain of interest for nonlinear function approximations where dutch traces are not available.

Source: Sutton and Barto, Reinforcement Learning, 2nd Edition