Nicholas Butko

Postdoctoral Fellow
Institute for Neural Computation
Univ. of California at San Diego

Exploring Exploring since 2004

Jump to: Curriculum VitaePublicationsResearch InterestsPersonal

Curriculum Vitae

Updated 24 May, 2011.

Thesis, Publications, Tech Reports, & Talks

NOTE: For PDF versions of papers, please see the MPLab publications page.

Research Interests

My ultimate interest is the grand quest of Cognitive Engineering, Viz. How can we build machines that are as truly adaptable and intelligent as organisms?

The question of how we can define, understand, and construct intelligence has been addressed by many great minds. In the beginning, it was thought that intelligence was ultimately rooted in reason and thought. My own experience and intuition suggest that this notion should be turned around. Reason and thought do not form a core from which perception and action spring and through which they mingle; perception and action underlie and give rise to everything cognitive.

It is my belief that we can only advance AI by commiting to approaches that are ultimately rooted in pereception, action, and iteraction with an uncertain and changing world. To this end, we must be students of probability, uncertainty, dynamics, information, and prediction. To address the problems of interacting with a changing world, it is important to commit ourselves to efficient methods that sacrifice "correctness" for getting the job done in time, before the world has moved on without us. We must create theories that are robust in the face of what the world can throw at us, and we must implement those theories in the real world (or domains of similar complexity) to make sure they hold.

A model of intelligence must prove itself by doing more intelligent things in the real world than researchers have been able to achieve before.



Classical Guitar: