Personalized Agents: AAAI Fall Symposium 2002


A personalized agent is idealized as an always-on autonomous program that acts as personal assistant for a specific user. The agent usually "lives" in a virtual world, has access to data about its user, and is empowered to act on its user's behalf in a variety of tasks. We believe that a key component to a successful personalized agent is learning from the user, both supervised and unsupervised. This symposium focused on the unique issues that arise when developing adaptive learning methods for such agents.


Post-Symposium Links


Original Links