Cognitivity & Intelligence
Creativity is inventing, experimenting, growing, taking risks, breaking rules, making mistakes, and having fun.
Mary Lou Cook
This section presents my pursuits and achievements in the area of cognitivity and intelligence, especially related to the intelligent behavior in machines. The page also provides a set of links that help to get deeper into the field of artificial intelligence.
My interests in artificial intelligence – or, I'd rather say, programmed intelligence – mostly concentrate on applying various intelligent techniques to modern video games and civil engineering systems I develop. Along with the complex design of these applications, I build up some combinations of hybrid expert systems and learning methods to achieve the real-life AI actions in a decision process. Utilized techniques include artificial neural networks, analytical processing and fuzzy controllers tuned with evolutionary algorithms. I'm also interested in general AI topics, intelligent houses, and nanotechnology.
Pursuits
Having developed the Neuritis shell expert system, I've got involved with the hybrid artificial intelligence solutions. I'm especially interested in splitting complex problems to smaller parts that can be efficiently solved by simple AI components, like multi-layered perceptrons or fuzzy logic controllers. I'm also assessing how these simple components can be trained using various optimization techniques, including evolutionary algorithms.
Game artificial intelligence refers to techniques used in computer and video games to produce the illusion of intelligence in the behavior of non-player characters (NPCs). The techniques used typically draw upon existing methods from the academic field of artificial intelligence (AI). However, the term game AI is often used to refer to a broad set of algorithms that also include techniques from control theory, robotics, computer graphics and computer science in general.
Working for a classified project, I did a lot of research on modern natural language processing, including functional and generative grammars, using statistical and logic-based approaches. I'm looking forward to further develop in that area, heading for perfect NLP solutions.
Achievements
The expert system shell for hybridizing artificial intelligence techniques: fuzzy logic controllers, neural networks, analytical components, and other user-defined elements.
The expert system for technical condition evaluation of railway bridge structures – currently used widely by the Polish State Railways. This application was created with the Neuritis package, using neural networks and analytical methods.
Challenging rule-based bots for our PC adaptation of the award-winning economic strategy board game Puerto Rico.
Links
Although Wikipedia should rather not be considered a scientific source at the moment, it's a good place to start your journey into the land of Artificial Intelligence. You'll find there the academic theory of AI (both within conventional and soft computing approach), AI publications, news, science-fiction and philosophical connotations.
CiteSeer is a scientific literature digital library and search engine that focuses primarily on the literature in computer and information science. CiteSeer aims to improve the dissemination and feedback of the scientific literature and to provide improvements in functionality, usability, availability, cost, comprehensiveness, efficiency, and timeliness in the access of scientific and scholarly knowledge.
Top image: A DNA sequence fragment, a Purkinje neuron cell, and a human brain model. Combined academic and public domain images.