Abstract
Automated Theory Formation is a hybrid AI technique which has been implemented in two scientific discovery systems, HR1 and HR2, both of which have been used successfully in various applications. We describe here the latest iteration in the HR series, in terms of the lessons learned from the successes and failures of the previous versions, and how these lessons have informed our design choices and the implementation details of the new version. We also present two case studies: a synthetic domain mirroring an aspect of medical diagnosis, and invariant discovery in formal methods. In each case, we compare HR3 with HR2 to highlight various improvements in the new version.
| Original language | English |
|---|---|
| Title of host publication | Proceedings of the 50th Anniversary Convention of the AISB |
| Editors | Roger Kibble |
| Place of Publication | London UK |
| Publisher | Goldsmiths |
| Number of pages | 8 |
| Publication status | Published - 2014 |
| Externally published | Yes |
| Event | AISB Convention (Society for the Study of Artificial Intelligence and Simulation of Behaviour) 2014 - London, United Kingdom Duration: 1 Apr 2014 → 4 Apr 2014 Conference number: 50th http://aisb50.org (Websit) |
Conference
| Conference | AISB Convention (Society for the Study of Artificial Intelligence and Simulation of Behaviour) 2014 |
|---|---|
| Abbreviated title | AISB 2014 |
| Country/Territory | United Kingdom |
| City | London |
| Period | 1/04/14 → 4/04/14 |
| Internet address |
|