Learning Abstract World Model for Value-preserving Planning with Options

TL;DR

This paper introduces a method for learning abstract world models that enable value-preserving planning with options, improving decision-making efficiency in complex, sensor-rich environments by operating at higher abstraction levels.

Contribution

It presents a novel approach to learn abstract MDPs from sensorimotor experiences, ensuring bounded value loss during planning with temporally-extended actions.

Findings

Abstract model learning enhances sample efficiency in planning.

Planning with abstract MDPs achieves successful goal navigation.

The approach maintains bounded value loss in original MDPs.

Abstract

General-purpose agents require fine-grained controls and rich sensory inputs to perform a wide range of tasks. However, this complexity often leads to intractable decision-making. Traditionally, agents are provided with task-specific action and observation spaces to mitigate this challenge, but this reduces autonomy. Instead, agents must be capable of building state-action spaces at the correct abstraction level from their sensorimotor experiences. We leverage the structure of a given set of temporally-extended actions to learn abstract Markov decision processes (MDPs) that operate at a higher level of temporal and state granularity. We characterize state abstractions necessary to ensure that planning with these skills, by simulating trajectories in the abstract MDP, results in policies with bounded value loss in the original MDP. We evaluate our approach in goal-based navigation environments that require continuous abstract states to plan successfully and show that abstract model learning improves the sample efficiency of planning and learning.