Skip to yearly menu bar Skip to main content


( events)   Timezone:  
Poster
Wed Dec 11 05:00 PM -- 07:00 PM (PST) @ East Exhibition Hall B + C #207
Loaded DiCE: Trading off Bias and Variance in Any-Order Score Function Gradient Estimators for Reinforcement Learning
Gregory Farquhar · Shimon Whiteson · Jakob Foerster
[ Paper [ Slides

Gradient-based methods for optimisation of objectives in stochastic settings with unknown or intractable dynamics require estimators of derivatives. We derive an objective that, under automatic differentiation, produces low-variance unbiased estimators of derivatives at any order. Our objective is compatible with arbitrary advantage estimators, which allows the control of the bias and variance of any-order derivatives when using function approximation. Furthermore, we propose a method to trade off bias and variance of higher order derivatives by discounting the impact of more distant causal dependencies. We demonstrate the correctness and utility of our estimator in analytically tractable MDPs and in meta-reinforcement-learning for continuous control.