RAPTOR: End-to-end Risk-Aware MDP Planning and Policy Learning by Backpropagation

TL;DR

RAPTOR introduces a risk-aware planning framework that optimizes the entropic utility in stochastic MDPs using backpropagation, enabling risk-sensitive decision-making in complex environments.

Contribution

It presents a novel reparameterization technique allowing end-to-end risk-aware optimization via backpropagation in stochastic environments.

Findings

Successfully applied to navigation, HVAC, and reservoir control domains.

Demonstrates effective risk management in highly stochastic MDPs.

Outperforms risk-agnostic methods in complex scenarios.

Abstract

Planning provides a framework for optimizing sequential decisions in complex environments. Recent advances in efficient planning in deterministic or stochastic high-dimensional domains with continuous action spaces leverage backpropagation through a model of the environment to directly optimize actions. However, existing methods typically not take risk into account when optimizing in stochastic domains, which can be incorporated efficiently in MDPs by optimizing the entropic utility of returns. We bridge this gap by introducing Risk-Aware Planning using PyTorch (RAPTOR), a novel framework for risk-sensitive planning through end-to-end optimization of the entropic utility objective. A key technical difficulty of our approach lies in that direct optimization of the entropic utility by backpropagation is impossible due to the presence of environment stochasticity. The novelty of RAPTOR lies in the reparameterization of the state distribution, which makes it possible to apply stochastic backpropagatation through sufficient statistics of the entropic utility computed from forward-sampled trajectories. The direct optimization of this empirical objective in an end-to-end manner is called the risk-averse straight-line plan, which commits to a sequence of actions in advance and can be sub-optimal in highly stochastic domains. We address this shortcoming by optimizing for risk-aware Deep Reactive Policies (RaDRP) in our framework. We evaluate and compare these two forms of RAPTOR on three highly stochastic do-mains, including nonlinear navigation, HVAC control, and linear reservoir control, demonstrating the ability to manage risk in complex MDPs.