Towards Software Apprentices that Learn in Dynamic Domains
Abstract
This project made progress on how to create software apprentices, intelligent systems that can operate as collaborators. We did this by developing an account of the representations needed to support flexible learning about dynamic systems, investigating the language understanding and sketch understanding capabilities needed to enable intelligent systems to interact with people as apprentices do, and exploring how to build more autonomous systems, capable of managing their own learning capabilities. This final report describes our progress. We summarize our work on autonomy (including learning via experimentation), natural language, qualitative decision-making, learning from episodic memories, and sketch recognition.
Document Details
- Document Type
- Technical Report
- Publication Date
- Aug 17, 2021
- Accession Number
- AD1146068
Entities
People
- Ken Forbus
Organizations
- Northwestern University