DeepAI AI Chat
Log In Sign Up

Knowledge-Grounded Reinforcement Learning

by   Zih-Yun Chiu, et al.
The Regents of the University of California
University of California, San Diego

Receiving knowledge, abiding by laws, and being aware of regulations are common behaviors in human society. Bearing in mind that reinforcement learning (RL) algorithms benefit from mimicking humanity, in this work, we propose that an RL agent can act on external guidance in both its learning process and model deployment, making the agent more socially acceptable. We introduce the concept, Knowledge-Grounded RL (KGRL), with a formal definition that an agent learns to follow external guidelines and develop its own policy. Moving towards the goal of KGRL, we propose a novel actor model with an embedding-based attention mechanism that can attend to either a learnable internal policy or external knowledge. The proposed method is orthogonal to training algorithms, and the external knowledge can be flexibly recomposed, rearranged, and reused in both training and inference stages. Through experiments on tasks with discrete and continuous action space, our KGRL agent is shown to be more sample efficient and generalizable, and it has flexibly rearrangeable knowledge embeddings and interpretable behaviors.


page 1

page 2

page 3

page 4


Godot Reinforcement Learning Agents

We present Godot Reinforcement Learning (RL) Agents, an open-source inte...

Newtonian Action Advice: Integrating Human Verbal Instruction with Reinforcement Learning

A goal of Interactive Machine Learning (IML) is to enable people without...

Opponent Aware Reinforcement Learning

In several reinforcement learning (RL) scenarios such as security settin...

Asking for Knowledge: Training RL Agents to Query External Knowledge Using Language

To solve difficult tasks, humans ask questions to acquire knowledge from...

Modeling Human Cognition with a Hybrid Deep Reinforcement Learning Agent

Human cognition model could help us gain insights in how human cognition...

Learning to act: a Reinforcement Learning approach to recommend the best next activities

The rise of process data availability has led in the last decade to the ...

A new soft computing method for integration of expert's knowledge in reinforcement learn-ing problems

This paper proposes a novel fuzzy action selection method to leverage hu...