Multi-task Maximum Entropy Inverse Reinforcement Learning

05/22/2018
by   Adam Gleave, et al.
0

Multi-task Inverse Reinforcement Learning (IRL) is the problem of inferring multiple reward functions from expert demonstrations. Prior work, built on Bayesian IRL, is unable to scale to complex environments due to computational constraints. This paper contributes the first formulation of multi-task IRL in the more computationally efficient Maximum Causal Entropy (MCE) IRL framework. Experiments show our approach can perform one-shot imitation learning in a gridworld environment that single-task IRL algorithms require hundreds of demonstrations to solve. Furthermore, we outline how our formulation can be applied to state-of-the-art MCE IRL algorithms such as Guided Cost Learning. This extension, based on meta-learning, could enable multi-task IRL to be performed for the first time in high-dimensional, continuous state MDPs with unknown dynamics as commonly arise in robotics.

READ FULL TEXT
research
03/22/2022

A Primer on Maximum Causal Entropy Inverse Reinforcement Learning

Inverse Reinforcement Learning (IRL) algorithms infer a reward function ...
research
04/27/2020

Maximum Entropy Multi-Task Inverse RL

Multi-task IRL allows for the possibility that the expert could be switc...
research
07/14/2021

Deep Adaptive Multi-Intention Inverse Reinforcement Learning

This paper presents a deep Inverse Reinforcement Learning (IRL) framewor...
research
12/01/2020

Revisiting Maximum Entropy Inverse Reinforcement Learning: New Perspectives and Algorithms

We provide new perspectives and inference algorithms for Maximum Entropy...
research
07/26/2019

Learning Task Specifications from Demonstrations via the Principle of Maximum Causal Entropy

In many settings (e.g., robotics) demonstrations provide a natural way t...
research
03/22/2022

X-MEN: Guaranteed XOR-Maximum Entropy Constrained Inverse Reinforcement Learning

Inverse Reinforcement Learning (IRL) is a powerful way of learning from ...
research
08/12/2020

Learning to Reason in Round-based Games: Multi-task Sequence Generation for Purchasing Decision Making in First-person Shooters

Sequential reasoning is a complex human ability, with extensive previous...

Please sign up or login with your details

Forgot password? Click here to reset