META-Learning Eligibility Traces for More Sample Efficient Temporal Difference Learning

by   Mingde Zhao, et al.

Temporal-Difference (TD) learning is a standard and very successful reinforcement learning approach, at the core of both algorithms that learn the value of a given policy, as well as algorithms which learn how to improve policies. TD-learning with eligibility traces provides a way to do temporal credit assignment, i.e. decide which portion of a reward should be assigned to predecessor states that occurred at different previous times, controlled by a parameter λ. However, tuning this parameter can be time-consuming, and not tuning it can lead to inefficient learning. To improve the sample efficiency of TD-learning, we propose a meta-learning method for adjusting the eligibility trace parameter, in a state-dependent manner. The adaptation is achieved with the help of auxiliary learners that learn distributional information about the update targets online, incurring roughly the same computational complexity per step as the usual value learner. Our approach can be used both in on-policy and off-policy learning. We prove that, under some assumptions, the proposed method improves the overall quality of the update targets, by minimizing the overall target error. This method can be viewed as a plugin which can also be used to assist prediction with function approximation by meta-learning feature (observation)-based λ online, or even in the control case to assist policy improvement. Our empirical evaluation demonstrates significant performance improvements, as well as improved robustness of the proposed algorithm to learning rate variation.


page 1

page 2

page 3

page 4


A Greedy Approach to Adapting the Trace Parameter for Temporal Difference Learning

One of the main obstacles to broad application of reinforcement learning...

ProMP: Proximal Meta-Policy Search

Credit assignment in Meta-reinforcement learning (Meta-RL) is still poor...

Faster and More Accurate Learning with Meta Trace Adaptation

Learning speed and accuracy are of universal interest for reinforcement ...

A Unified Approach for Multi-step Temporal-Difference Learning with Eligibility Traces in Reinforcement Learning

Recently, a new multi-step temporal learning algorithm, called Q(σ), uni...

Preferential Temporal Difference Learning

Temporal-Difference (TD) learning is a general and very useful tool for ...

TBQ(σ): Improving Efficiency of Trace Utilization for Off-Policy Reinforcement Learning

Off-policy reinforcement learning with eligibility traces is challenging...

Improving the Efficiency of Off-Policy Reinforcement Learning by Accounting for Past Decisions

Off-policy learning from multistep returns is crucial for sample-efficie...

Please sign up or login with your details

Forgot password? Click here to reset