Stochastic grammar based incremental machine learning using scheme
Date
Authors
Editor(s)
Advisor
Supervisor
Co-Advisor
Co-Supervisor
Instructor
BUIR Usage Stats
views
downloads
Citation Stats
Series
Abstract
Gigamachine is our initial implementation of an Artificial General Intelligence (AGI system) in the O'Caml language with the goal of building Solomonoff's Phase 1 machine that he proposed as the basis of a quite powerful incremental machine learning system (Sol02). While a lot of work remains to implement the full system, the present algorithms and implementation demonstrate the issues in building a realistic system. Thus, we report on our ongoing research to share our experience in designing such a system. In this extended abstract, we give an overview of our present implementation, summarize our contributions, discuss the results obtained, the limitations of our system, our plans to overcome those limitations, potential applications, and future work. The reader is referred to (Sch04; Sol02; Sol09) for a background on general-purpose incremental machine learning. The precise technical details of our ongoing work may be found in (Ozk09), which focuses on our algorithmic contributions. The discussion here is not as technical but assumes basic knowledge of universal problem solvers.
Gigamachine is our initial implementation of an Artificial General Intelligence (AGI system) in the O'Caml language with the goal of building Solomonoff's Phase 1 machine that he proposed as the basis of a quite powerful incremental machine learning system (Sol02). While a lot of work remains to implement the full system, the present algorithms and implementation demonstrate the issues in building a realistic system. Thus, we report on our ongoing research to share our experience in designing such a system. In this extended abstract, we give an overview of our present implementation, summarize our contributions, discuss the results obtained, the limitations of our system, our plans to overcome those limitations, potential applications, and future work. The reader is referred to (Sch04; Sol02; Sol09) for a background on general-purpose incremental machine learning. The precise technical details of our ongoing work may be found in (Ozk09), which focuses on our algorithmic contributions. The discussion here is not as technical but assumes basic knowledge of universal problem solvers.