AI in Networks

Course Curriculum

1. Evolution of networks: Circuit switching, packet switching, TCP/ IP, VoIP, Network layers; 

2. Need for AI in Networks: Applications of AI in different network layers

3. Statistical learning: KL-divergence minimization, Empirical risk minimization, Bayes classifier, ML-MAP estimator, Bias-variance tradeoff, Regularization, Supervised and unsupervised learning, 

4. Parameter Optimization: Convex optimization fundamentals, Lagrangian-dual formulation, KKT conditions, linear programing, quadratic programming, gradient descent, sub-gradient descent, Non-convex optimization algorithms: Branch and bound method, simulated annealing, genetic algorithms, ant colony optimization

5. Use cases of AI in networks: Wireless receiver design, routing in networks, tracking a mobile object in a network (programming exercises)

Grading 

Assignments/ Quiz: 40%

Programming exercises: 30% 

Final exam: 30%