Agent Architectures for structured uncertain environments
Zahra Zamani (SoCS CECS)
CS HDR MONITORING AI Research GroupDATE: 2010-04-16
TIME: 09:30:00 - 10:00:00
LOCATION: RSISE Seminar Room, ground floor, building 115, cnr. North and Daley Roads, ANU
CONTACT: JavaScript must be enabled to display this email address.
ABSTRACT:
My area of research is to develop architectures for agents operating in structured uncertain environments. The agents will receive percepts from the environment in a structured format, based on higher-order knowledge representation. It will then learn to act optimally in an environment with uncertainty.
The features of the state space will be represented in a rich higher-order Modal language. This will provide high expressive knowledge for every state of the world with the probability associated with the uncertainty of the agent's knowledge about the states. Because of uncertainty the agent will have a belief over the possible states it may be at a certain time. We will use the motion-sensor framework in the probabilistic robotics context to make sure that the uncertainty on the states is handled.
As the goal of the agent is to act optimally based on the long-run expected total rewards, action selection methods suitable for our architecture will be used and the agent has to learn how to act optimally in such environment.
BIO:
PhD student in Computer Science.
