Reasoning about Rational Agents from a Programming Languages Perspective. Rational agents are complex system components inspired by models of rationality, in which beliefs, desires and intentions are represented explicitly. Intelligent agents are fast becoming mainstream and have wide-ranging industrial application. However, a serious obstacle to the deployment of rational agents is a lack of understanding of how to design and reason about agents to provide performance and correctness guarantees ....Reasoning about Rational Agents from a Programming Languages Perspective. Rational agents are complex system components inspired by models of rationality, in which beliefs, desires and intentions are represented explicitly. Intelligent agents are fast becoming mainstream and have wide-ranging industrial application. However, a serious obstacle to the deployment of rational agents is a lack of understanding of how to design and reason about agents to provide performance and correctness guarantees. This project will investigate extensions to intelligent agent architectures and develop tools for reasoning about complex systems built using those architectures. This will not only enable reliable prediction of performance, but contribute to a general methodology for intelligent agent design.
Read moreRead less
RichProlog, a System for Deducing, Inducing and Learning in the Declarative Programming Paradigm. The aim of the project is to contribute to bridge the gap between learning and logic, theoretically and practically. Our purpose is to extend considerably the scope of the declarative programming paradigm, and build a system that can be used to solve learning or discovery problems as encountered in Artificial Intelligence. The system will enable rapid prototyping when applied to problems involving d ....RichProlog, a System for Deducing, Inducing and Learning in the Declarative Programming Paradigm. The aim of the project is to contribute to bridge the gap between learning and logic, theoretically and practically. Our purpose is to extend considerably the scope of the declarative programming paradigm, and build a system that can be used to solve learning or discovery problems as encountered in Artificial Intelligence. The system will enable rapid prototyping when applied to problems involving deduction, induction, and nonmonotonic reasoning. We intend the system to become a standard tool for tackling a broad range of applications, and the underlying theory to provide new insights on the logical foundations of Artificial Intelligence.
Read moreRead less
Structures and Protocols for Inference. The proposed research is expected to lead to increased adoption and efficiency of use of machine learning technologies. It will develop new and better ways to use existing machine learning software in a manner that allows easier integration into commercial products. It will increase the competitiveness of Australian industry.
Foundations of Executable Temporal Logic. In many computer applications, including those of temporal reasoning, distributed computations and knowledge representations, the concept of time is of central importance. Multiple granularity of time also plays a critical role as not all events are necessarily defined over a uniform model of time. This project will develop the foundations of executable logical representations, supporting multiple granularity of time. This will allow system developers a ....Foundations of Executable Temporal Logic. In many computer applications, including those of temporal reasoning, distributed computations and knowledge representations, the concept of time is of central importance. Multiple granularity of time also plays a critical role as not all events are necessarily defined over a uniform model of time. This project will develop the foundations of executable logical representations, supporting multiple granularity of time. This will allow system developers access to powerful logical techniques in those applications. In the process, fundamental problems in modelling multiple granularity of time will be identified, and application-independent solutions to those problems will be provided.Read moreRead less