Learning Manner of Execution from Partial Corrections

TL;DR

This paper presents a model enabling an agent to learn appropriate action execution manners in various contexts through trial, error, and verbal corrections, facilitating effective symbol grounding for context-aware planning.

Contribution

The paper introduces a novel approach where an agent learns context-specific action manners from partial corrections without prior concept knowledge.

Findings

Agent successfully learns context-dependent action manners.

Model effectively grounds symbols from verbal feedback.

Approach improves context-aware action execution.

Abstract

Some actions must be executed in different ways depending on the context. For example, wiping away marker requires vigorous force while wiping away almonds requires more gentle force. In this paper we provide a model where an agent learns which manner of action execution to use in which context, drawing on evidence from trial and error and verbal corrections when it makes a mistake (e.g., ``no, gently''). The learner starts out with a domain model that lacks the concepts denoted by the words in the teacher's feedback; both the words describing the context (e.g., marker) and the adverbs like ``gently''. We show that through the the semantics of coherence, our agent can perform the symbol grounding that's necessary for exploiting the teacher's feedback so as to solve its domain-level planning problem: to perform its actions in the current context in the right way.