This course introduces students to the basic knowledge representation, problem solving, and learning methods of artificial intelligence. Upon completion of 6.034, students should be able to develop intelligent systems by assembling solutions to concrete computational problems, understand the role of knowledge representation, problem solving, and learning in intelligent-system engineering, and appreciate the role of problem solving, vision, and language in understanding human intelligence from a computational perspective.

Artificial Intelligence : Course lectures, hours 42. There are 11 pdf files, total 562 pages. Topics : Introduction to AI; Problem Solving - Search and Control Strategies; Knowledge Representation Issues - Predicate Logic, Rules; Reasoning System - Symbolic, Statistical; Game Playing; Learning; Expert System; Fundamentals of Neural Networks; Fundamentals of Genetic Algorithms; Natural Language Processing; Common Sense.

Soft Computing : Course lectures, hours 42. There are 9 pdf files, total 398 pages. Topics : Introduction to Soft Computing; Fundamentals of Neural Network; Back Propagation Network; Associative Memory; Adaptive Resonance Theory; Fuzzy Set Theory; Fuzzy Systems; Fundamentals of Genetic Algorithms; Hybrid Systems.

Introduces representations, techniques, and architectures used to build applied systems and to account for intelligence from a computational point of view. Applications of rule chaining, heuristic search, constraint propagation, constrained search, inheritance, and other problem-solving paradigms. Applications of identification trees, neural nets, genetic algorithms, and other learning paradigms. Speculations on the contributions of human vision and language systems to human intelligence.

There is need for a rigorous, quantitative multidisciplinary design methodology that works with the non-quantitative and creative side of the design process in engineering systems. The goal of multidisciplinary systems design optimization is to create advanced and complex engineering systems that must be competitive not only in terms of performance, but also in terms of life-cycle value. The objective of the course is to present tools and methodologies for performing system optimization in a multidisciplinary design context. Focus will be equally strong on all three aspects of the problem: (i) the multidisciplinary character of engineering systems, (ii) design of these complex systems, and (iii) tools for optimization.

Neural structures and mechanisms mediating the detection, localization, and recognition of sounds. Discussion of how acoustic signals are coded by auditory neurons, the impact of these codes on behavorial performance, and the circuitry and cellular mechanisms underlying signal transformations. Topics include temporal coding, neural maps and feature detectors, learning and plasticity, and feedback control. General principles are conveyed by theme discussions of auditory masking, sound localization, musical pitch, speech coding, and cochlear implants, and auditory scene analysis.

Inspired by reality-based computing from the natural world, this course covers several unconventional computational methods and theories, such as quantum computation, DNA and molecular computation, genetic algorithms, self-organizing networks, and cellular automata. Note: for this course, it will be quite helpful to have a working knowledge of cellular biology (available from the Saylor Foundation’s BIO301). Upon successful completion of this course, the student will be able to: describe abstracted finite-memory program, a finite state automaton, and regular language; list and explain the characteristics of universal Turing transducers; describe the computational idea behind the DNA-based computer; explain the differences between bio-electronic, biochemical, and biomechanical computers; describe the functional principles of genetic algorithms and list their limitations; define the cellular automaton and the cellular neural network, and show examples of how they compute; describe how logic gates may be constructed for quantum bits; describe a simple model for a quantum computer based on a classical computer; describe an algorithm which makes use of quantum parallelism. This free course may be completed online at any time. (Computer Science 411)

Fundamentals of characterizing and recognizing patterns and features of interest in numerical data. Basic tools and theory for signal understanding problems with applications to user modeling, affect recognition, speech recognition and understanding, computer vision, physiological analysis, and more. Decision theory, statistical classification, maximum likelihood and Bayesian estimation, non-parametric methods, unsupervised learning and clustering. Additional topics on machine and human learning from active research.

This course surveys a variety of reasoning, optimization, and decision-making methodologies for creating highly autonomous systems and decision support aids. The focus is on principles, algorithms, and their applications, taken from the disciplines of artificial intelligence and operations research. Reasoning paradigms include logic and deduction, heuristic and constraint-based search, model-based reasoning, planning and execution, reasoning under uncertainty, and machine learning. Optimization paradigms include linear, integer and dynamic programming. Decision-making paradigms include decision theoretic planning, and Markov decision processes. This course is offered both to undergraduate (16.410) students as a professional area undergraduate subject, in the field of aerospace information technology, and graduate (16.413) students.