Minds & Machines - Artificial Intelligence

Artificial Intelligence is the attempt to artifically create cognitive beings. AI researchers like to point out that the cognitive powers of these beings need not be the same as humans, but as traditional Artificial Intelligence techniques have fallen far short of the overly optimistic predictions of 'having thinking machines within a few decades' - and have not produced a robot that is even remotely like the robots of the science fiction movies - many AI researchers have turned to the cognitive psychologists for inspiration: a clear vindication of the need of cognitive science, bringing together cognitive psychology and artificial intelligence. Moreover, while artificial intelligence has an obvious engineering flavor to it, it also has a highly theoretical component: cognitive science's attempt to map out and understand all of cognition, human or otherwise. Finally, artificial intelligence raises deep philosophical questions: do thinking machines have personal rights? Do we really want to create artificial intelligence? Is it even possible? What, indeed, are the basic foundations of cognition in general?

Below are 4 subfields of Artificial Intelligence that are well supported by Minds & Machinesin terms of courses, faculty, and laboratories:

Machine Reasoning

Machine Reasoning is the attempt to have a computer do, or be an aid with, reasoning. Historically, the main focus of machine reasoning has been Automated Theorem Proving, where one attempts of formalize algorithms to check whether something deductively follows from something else. However, organisms trying to reason about an uncertain and dynamic environment would probably use reasoning to generate hypotheses, or make non-deductive, yet quite reasonable, extrapolations from what they know.

Recommended Dual Major: CSCI/PHIL, CSCI/PSYC
Recommended Courses: Introduction to Cognitive Science, Introduction to Logic, Intermediate Logic, Philosophy of AI, Logic and AI
Associated Faculty: Selmer Bringsjord, Nick Cassimatis, Bram van Heuveln
Associated Labs: RAIR Lab
top

Cognitive Robotics

Cognitive Robotics is the attempt to create robots with a wide spectrum of cognitive powers, as needed for the kinds of open-ended tasks and problems that any organism in a sufficiently complex world will face. Cognitive robotics is in this respect to be distinguished from more traditional industrial robotics where the task to be performed is narrow and well-defined and the environment well-contained.

Recommended Dual Major: CSE/PSYC
Recommended Courses: Introduction to Cognitive Science
Associated Faculty: Selmer Bringsjord, Nick Cassimatis, Bram van Heuveln
Associated Labs: RAIR Lab

top

Synthetic Characters

Synthetic characters are agents that live in virtual or cybernetic environments such as video games or the internet. Given their more narrow environment, a synthetic characters' psychology can be a lot simpler than a cognitive being having to navigate our actual physical world. At the same time, however, their desired or needed cognitive powers can still stump the best of programmers, which is why a firm grounding in cognitive science is needed for this. Clearly, the gaming inductry is interested in having game characters whose cognitive model is a little more intricate than the sliding personality scales of the Sims, but other applications would be intelligent tutoring systems or more realistic virtual environments used for simulation and training purposes.

Recommended Dual Major: CSCI/PSYC
    See also: Minor in Game Studies
Recommended Courses: Introduction to Cognitive Science
Associated Faculty: Selmer Bringsjord, Marc Destefano, Wayne Gray, Nick Cassimatis
Associated Labs: CogWorks Lab, RAIR Lab
top

Philosophy of AI

The field of Artificial Intelligence is rife with deep philosophical questions. For example, can the mind be captured computationally? Do the Godel Results mean that it can't? What is consciousness? What constitues a 'personal being'? If mental activity is nothing but computation, is there free will? Or moral responsibility? Or rights? Philosophers of AI also debate the plausibility of different models of cognition, such as Logicism, Connectionism, and Situated Cognition, and particular problems that have come up during the lifetime of AI, such as the Chinese Room argument and the Turing Test, the grounding problem and the frame problem.

Recommended Dual Major: CSCI/PHIL
Recommended Courses: Introduction to Philosophy, Introduction to Cognitive Science, Introduction to Logic, Knowledge, Belief and Cognition, Philosophy of AI, Computability and Logic
Associated Faculty: Selmer Bringsjord, Jim Fahey, Bram van Heuveln
Associated Labs: RAIR Lab
top

For more information contact: Bram van Heuveln