| Course Code: CS6398 Course Name: Nature-Inspired Computing Syllabus: Nature-Inspired Computing encompasses computational paradigms that draw inspiration from biological, ecological, and physical processes observed in nature. Such methods have found their way into a plethora of application areas including optimization, machine learning, decentralized and distributed control, robotics and the modelling of complex systems. While many contemporary Artificial Intelligence (AI) techniques rely heavily on large datasets and models based on variants of Artificial Neural Networks, it is increasingly important to explore alternative and complementary paradigms that can augment, enrich and refine these approaches. In natural systems, many processes - despite their inherent complexity - are carried out through population-based interactions, where individual entities perform sensorimotor toil and continuously exchange information with one another, collectively giving rise to emergent intelligence and adaptive behavior. This course aims to encourage students to think out-of-the-box, beyond conventional computational frameworks, fostering an appreciation of the subtle yet sophisticated mechanisms through which nature tackles and solves complex problems. The course is intended to motivate students to think unconventionally, analyze, adapt, and innovate upon these natural principles and design novel computational solutions. Content: Introduction to Nature-Inspired Computing: Limitations of classical optimization techniques, Motivation for nature-inspired algorithms, Biological inspirations in computation, Populationbased vs trajectory-based methods, Exploration vs exploitation, Fitness landscapes, Applications in engineering, AI, robotics and networks. Evolutionary Computation: Genetic Algorithms (GA): Biological foundations of evolution, Representation and encoding, Fitness evaluation, Selection methods, Crossover and Mutation operators, Elitism and Replacement strategies, Convergence analysis, Genetic programming basics, Applications of GA to optimization, learning, Schema theorem. Swarm Intelligence: Collective behaviors in biological systems, Social behaviors in shoals and flocks; Particle Swarm Optimization (PSO): Velocity and position updates, Local/global bests, Convergence, Parameter selection and Applications; Ant Colony Optimization (ACO): Foraging behavior in ants, Pheromone trails, Probabilistic path selection, Ant system and variants, Applications to Routing problems, Mobile robotics. Artificial Immune Systems (AIS): Biological Immune System (BIS) overview, Self vs Non-self discrimination, Theories - Clonal selection, Negative selection algorithm, Immune/Idiotypic network theory, Danger theory, Applications to Robotics and Pattern recognition. Other Nature-inspired Algorithms and Learning Systems: Bacterial foraging optimization, Cuckoo search; Multi-Agent Systems, Mobile Agents for decentralized and distributed applications, Reinforcement Learning and Federation, Applications to Swarm robotics. Laboratory Component: Students will be guided in applying these algorithms to emulate and solve real-world problems in both centralized and decentralized settings, and be made to systematically evaluate their performances. The assignments will involve the use of mobile agents and will also aim to connect and/or augment these agents and algorithms with state-of the-art AI techniques. Texts: 1. Xin-She Yang (2020). Nature-Inspired Optimization Algorithms (2 ed.). Academic Press 2. James F. Kennedy, Russell Eberhart & Yuhui Shi (2001). Swarm Intelligence. Morgan Kaufmann Series in Artificial Intelligence. Morgan Kaufmann 3. Andries P. Engelbrecht (2007). Computational Intelligence: An Introduction (2 ed.). John Wiley \& Sons 4. Leandro Nunes Castro & Jonathan Timmis (2002). Artificial Immune Systems: A New Computational Intelligence Approach. Springer |