Log In Sign Up

A Composable Specification Language for Reinforcement Learning Tasks

by   Kishor Jothimurugan, et al.

Reinforcement learning is a promising approach for learning control policies for robot tasks. However, specifying complex tasks (e.g., with multiple objectives and safety constraints) can be challenging, since the user must design a reward function that encodes the entire task. Furthermore, the user often needs to manually shape the reward to ensure convergence of the learning algorithm. We propose a language for specifying complex control tasks, along with an algorithm that compiles specifications in our language into a reward function and automatically performs reward shaping. We implement our approach in a tool called SPECTRL, and show that it outperforms several state-of-the-art baselines.


page 1

page 2

page 3

page 4


Compositional Reinforcement Learning from Logical Specifications

We study the problem of learning control policies for complex tasks give...

Reinforcement Learning Agent Training with Goals for Real World Tasks

Reinforcement Learning (RL) is a promising approach for solving various ...

Interactively Learning to Summarise Timelines by Reinforcement Learning

Timeline summarisation (TLS) aims to create a time-ordered summary list ...

Discovering Generalizable Spatial Goal Representations via Graph-based Active Reward Learning

In this work, we consider one-shot imitation learning for object rearran...

Direct Behavior Specification via Constrained Reinforcement Learning

The standard formulation of Reinforcement Learning lacks a practical way...

An Extensible Interactive Interface for Agent Design

In artificial intelligence, we often specify tasks through a reward func...

A Policy Search Method For Temporal Logic Specified Reinforcement Learning Tasks

Reward engineering is an important aspect of reinforcement learning. Whe...

Code Repositories


Reinforcement learning using logical specifications

view repo