SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2201.12126 · IJCAI · 2022

Leveraging Class Abstraction for Commonsense Reinforcement Learning via Residual Policy Gradient Methods

Niklas Höpner, Herke Van Hoof, Ilaria Tiddi

arXiv · PDF · Open in the Atlas

Code that ran

We have not lifted any functions out of this paper's repositories yet, so there is nothing we have run. The repositories linked to it are listed below.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Enabling reinforcement learning (RL) agents to leverage a knowledge base while learning from experience promises to advance RL in knowledge intensive domains. However, it has proven difficult to leverage knowledge that is not manually tailored to the environment. We propose to use the subclass relationships present in open-source knowledge graphs to abstract away from specific objects. We develop a residual policy gradient method that is able to integrate knowledge across different abstraction levels in the class hierarchy. Our method results in improved sample efficiency and generalisation to unseen objects in commonsense games, but we also investigate failure modes, such as excessive noise in the extracted class knowledge or environments with little class structure.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2201.12126")
get_code_for_paper("2201.12126")
have("2201.12126")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2201.12126.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2201.12126.svg)](https://syntology.ai/paper/2201.12126)

Connect an agent — have() is free.