DeepAI AI Chat
Log In Sign Up

An Analysis of Measure-Valued Derivatives for Policy Gradients

by   João Carvalho, et al.
Technische Universität Darmstadt

Reinforcement learning methods for robotics are increasingly successful due to the constant development of better policy gradient techniques. A precise (low variance) and accurate (low bias) gradient estimator is crucial to face increasingly complex tasks. Traditional policy gradient algorithms use the likelihood-ratio trick, which is known to produce unbiased but high variance estimates. More modern approaches exploit the reparametrization trick, which gives lower variance gradient estimates but requires differentiable value function approximators. In this work, we study a different type of stochastic gradient estimator - the Measure-Valued Derivative. This estimator is unbiased, has low variance, and can be used with differentiable and non-differentiable function approximators. We empirically evaluate this estimator in the actor-critic policy gradient setting and show that it can reach comparable performance with methods based on the likelihood-ratio or reparametrization tricks, both in low and high-dimensional action spaces. With this work, we want to show that the Measure-Valued Derivative estimator can be a useful alternative to other policy gradient estimators.


page 1

page 2

page 3

page 4


An Empirical Analysis of Measure-Valued Derivatives for Policy Gradients

Reinforcement learning methods for robotics are increasingly successful ...

Total stochastic gradient algorithms and applications in reinforcement learning

Backpropagation and the chain rule of derivatives have been prominent; h...

Do Differentiable Simulators Give Better Policy Gradients?

Differentiable simulators promise faster computation time for reinforcem...

Policy Gradient using Weak Derivatives for Reinforcement Learning

This paper considers policy search in continuous state-action reinforcem...

All-Action Policy Gradient Methods: A Numerical Integration Approach

While often stated as an instance of the likelihood ratio trick [Rubinst...

Enabling First-Order Gradient-Based Learning for Equilibrium Computation in Markets

Understanding and analyzing markets is crucial, yet analytical equilibri...

Causal Policy Gradients

Policy gradient methods can solve complex tasks but often fail when the ...