TECHNOLOGYReluctant Episodic Memory
Overview
Commonsense reasoning utilizes factual information, logic, and pattern recognition. The core capabilities of Reluctant Episodic Memory (REM) support commonsense reasoning with a memory-centric approach for agent behavior. The algorithm utilizes basic principles for remembering, predicting, and correlating new observations to alleviate the burden of modifying software as changes are introduced into an environment. Incomplete perception circuits formed by REM are repaired through continued exploration, additional training, or interactions with outside sources of knowledge such as PraxiNet. Analysis of such circuits as part of a systems design evaluation process will ensure that an autonomous agent can interact successfully and robustly with its environment.For additional information concerning the device and for information regarding licensing please contact the Office of Technology Commercialization. 301 405 3947 or by e-mail [email protected]
Contact Info
UM Ventures
0134 Lee Building
7809 Regents Drive
College Park, MD 20742
Email: [email protected]
Phone: (301) 405-3947 | Fax: (301) 314-9502