Course Information

Course Name	Turkish	Nöromorfik Hesaplama
	English	Neuromorphic Computing
Course Code	BLG 643	Credit	Lecture (hour/week)	Recitation (hour/week)	Laboratory (hour/week)
Semester	-	3	3	-	-
Course Language		English
Course Coordinator		Burak Berk Üstündağ
Course Objectives		1- Understanding neuromorphic computational paradigms of intelligent systems in nature 2- Understanding and performance measure of information processing in neuromorphic systems 3- Design of Spiking Neural Networks depending on neuron and neural system models 4- Design of neuromorphic systems for computation and machine learning 5- Implementation methods of neuromorphic information processing systems (Memristors,BNNs etc.)
Course Description		Due to development of AI applications, performing the cognitive functions by using nature inspired computational paradigms has become a major trend for increasing the power and data efficiency. Neuromorphic computing is based on investigation, modeling and emulation of biological neural systems and the brain-connectome structure. This course covers neuromorphic learning in brain-nerve-connectome structures, spiking neural networks and their artificial implementation, neuromorphic coding, basics of stochastic computing, biomimetic neural networks, cognitive functions and neuromorphic system applications.
Course Outcomes		M.Sc./Ph.D. students who successfully pass this course gain knowledge, skill and competency in the following subjects; 1- Understanding the requirements for changing the Von Neumann architecture depending on analogy between the neuromorphic systems in nature and the brain inspired artificial intelligence. 2- Understanding the reasoning, context awareness, situational awareness, and intelligence in cognitive systems and their test methods. 3- Modeling of stochastic coding relationship between structural formation and the processed signals in neuromorphic systems 4-Learning the basics of mathematical operations used in investigation and emulation of neuromorphic systems. 5-Understanding the neuron models and spiking neural networks (SNN) 1- Application of spiking neural networks (SNN) for computation and machine learning. 2- Application of Minimum Energy-Maximum Entropy principle in neuromorphic computing 3- Learning the recent physical implementation methods for neuromorphic computing (memristors etc.)
Pre-requisite(s)
Required Facilities
Other
Textbook		N.K. Kasabov, “Time-Space, Spiking Neural Networks and Brain-Inspired Artificial Intelligence”, Springer, 2019
Other References		Q. Yu, H. Tang, J. Hu, K.T. Chen, “Neuromorphic Cognitive Systems: A Learning and Memory Centered Approach”, Springer, 2017 M.Shahsavari, “Unconventional Computation From Digital to Brain-like Neuromorphic: Memristive Computing”, Springer, 2017 M.Suri, “Advances in Neuromorphic Hardware Exploiting Emerging Nanoscale Devices”, Springer, 2017 W.J.Gross, V.C.Gaduet “Stochastic Computing: Techniques and Applications”, Springer, 2019

Courses

Help

About