Skip to main content

IE 704: Selected Topics in A.I. for Operations Research

Prerequisite:  Instructor's permission

Contents

Overview of some problem types encountered in O.R. where A.I. approaches have proved useful: combinatorial optimization and Planning, clustering & pattern recognition, prediction and control of stochastic systems.

Constraint Satisfaction Problems; backtracking search, local search, intelligent backtracking. Strategies and algorithms for automated planning. Constraint directed scheduling.

Bayesian networks; simulation based algorithms for inference. Dynamic Bayesian networks; filtering and prediction; smoothing. Hidden Markov models. Kalman filters. Approximate inference in Dynamic Bayesian Networks; Particle Filtering. Learning Bayes Net structures and parameters. Learning with Hidden Variables. EM algorithm.

Feedforward neural networks and Radial Basis Function networks for function approximation.

Fuzzy systems for mapping; fuzzy associative memory and applications in control. Adaptive FAM and Neuro-fuzzy systems.

Learning Vector Quantization, Support Vector Machines, Self Organizing Feature Maps for pattern classification & recognition. Adaptive Resonance Theory networks.

Hopfield Network applications in combinatorial optimization. Simulated Annealing. Boltzmann Machine; applications in combinatorial optimization and pattern recognition.

Use of neural function approximations in neuro-dynamic programming and Reinforcement Learning of control policies.


References

 

  • S. Russell and P. Norvig, (2003), ‘Artificial Intelligence: A modern approach’. 2nd Edition, Pearson Education Asia
  • S. Haykin, (1999), ‘Neural Networks: A comprehensive foundation’, 2nd Ed., Pearson Education Asia.
  • L Fausett, (2004), ‘Fundamentals of Neural Networks: Architectures Algorithms and Applications’, Pearson Education.
  • J.S.R.Jang, C. T. Sun and E. Mizutani, ( 2004), ‘Neuro-Fuzzy and Soft Computing’, Pearson Education
  • B. Kosko , (1994), ‘Neural Networks and Fuzzy Systems: A dynamical systems approach to machine intelligence’, Prentice Hall India.
  • R.S. Sutton and A. G. Barto, (1998), ‘Reinforcement Learning: An Introduction’, MIT Press.