DeepAI
Log In Sign Up

A Composable Specification Language for Reinforcement Learning Tasks

08/21/2020
by   Kishor Jothimurugan, et al.
5

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.

READ FULL TEXT

page 1

page 2

page 3

page 4

06/25/2021

Compositional Reinforcement Learning from Logical Specifications

We study the problem of learning control policies for complex tasks give...
07/21/2021

Reinforcement Learning Agent Training with Goals for Real World Tasks

Reinforcement Learning (RL) is a promising approach for solving various ...
11/14/2022

Interactively Learning to Summarise Timelines by Reinforcement Learning

Timeline summarisation (TLS) aims to create a time-ordered summary list ...
11/24/2022

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

In this work, we consider one-shot imitation learning for object rearran...
12/22/2021

Direct Behavior Specification via Constrained Reinforcement Learning

The standard formulation of Reinforcement Learning lacks a practical way...
06/06/2019

An Extensible Interactive Interface for Agent Design

In artificial intelligence, we often specify tasks through a reward func...
09/27/2017

A Policy Search Method For Temporal Logic Specified Reinforcement Learning Tasks

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

Code Repositories

spectrl_tool

Reinforcement learning using logical specifications


view repo