This course considers neurological, psychological, and structural models of intelligence. It uses these models as a basis for discussion and development of new models that may exhibit potential for creating cognitive systems. The majority of biological intelligence creatures are simple, yet they can achieve complex information processing that current artificial intelligence cannot match. Can we use these simple models to learn how to design better artificial intelligence? Thus this course is a combination of what we know about intelligence with discovery what makes it possible.
The emphasis in this course is on the development of the concept of machine intelligence and architectures that may result in cognitive systems. The stress is on functional models of intelligence, perception, goal creation, motivations, attentions switching and cognition. We are interested in a full feature model of cognitive system, believing that only then proper perspective is used to develop perception and motor skills, to develop episodic and semantic memories, to have a focus of attention and to perform planning and thinking.
This subject lies at a cross-section of artificial intelligence, neural networks, neuroscience and cognitive psychology, and involves developing models that illustrate brain functions, observed cognitive phenomena and their behavioral manifestations. These models are used to develop embodied agents that interact with the environment through a physical body that is able to perceive and act on the environment.
Aims and objectives
The course aims to teach students about principles and structural organization of intelligence, learning and goal oriented behavior. Another aim is to study biological substrates underlying cognition, with focus on the neural models of mental processes and their behavioral manifestations. Rather that emulating the brain the course focuses on models of embodied intelligence that learns through interaction with environment.
The course addresses a number of issues important to development of cognitive architectures. It tries to define what it means to be intelligent, anticipate, learn from experience, make associations, perceive, act independently, be aware, switch attention and think. It discusses how the machine's interaction with its environment leads to better behavior, better understanding, and success of its mission. It points out the software and hardware issues in doing this efficiently and in real-time.
