SRI International, a non-profit research institute in Menlo Park, California, has made available a collection of successful machine learning and reasoning technologies developed on DARPA's Personalized Assistant that Learns (PAL) program.
The PAL research program focused on improving the way computers support humans through the use of cognitive systems that reason, learn from experience, and accept guidance in order to provide effective, personalized assistance.
The available technologies include both general-purpose learning methods along with more focused learning applications. Several of the technologies employ machine learning techniques to acquire models of user preferences and behaviors from both observed activities and explicit user feedback (e.g., C2RSS, CALO Express, iLink, Task Learning, Task Assistant, Time Manager).
The PAL capabilities have been modularized, packaged, and adapted to industry standards to facilitate their incorporation into target applications. Various infrastructure components and APIs are available to simplify interaction with the technologies.
These technologies were developed by research teams at SRI International, Carnegie Mellon University, the University of Massachusetts, the University of Rochester, the Institute for Human and Machine Cognition, Oregon State University, the University of Southern California, and Stanford University.