Basic Information

• Course Code: EE746
• Course Name: Neuromorphic Engineering
• Course Offered In: Autumn 2022-2023
• Instructors: Udayan Ganguly
• Prerequisites: None
• Difficulty (on a scale of 5): 4

Course Content

Introduction to Neuromorphic Engineering; Signalling and operation of Biological neurons, neuron models, signal encoding and statistics; Synapses and plasticity rules, biological neural circuits; Neuromorphic design principles; FETs – device physics and sub-threshold circuits; Analog and digital electronic neuron design; Non-volatile memristive semiconductor devices; Electronic synapse design; Interconnection Networks; Interconnection schemes for large non-spiking and spiking neural networks; Analysis of design, architecture and performance characteristics of demonstrated chips employing Analog neuromorphic VLSI, Digital neuromorphic VLSI, Electronic synapses and other neuromorphic systems.

Feedback on Lectures

Lectures hinged entirely on discussion and back-and-forth interaction between students and professor and amongst students as well. The content is fun and easy to follow if the student is consistent with their attendance.

Feedback on Evaluations

Evaluation was conducted in the form of 4 quizzes (which included midsem and endsem examinations), 3 assignments, and 1 project. The quiz content was based on class discussions and contained material that could only be understood by being there. The assignments were creation of models discussed in class using matlab or python. Assignments were lengthy but easy enough if one starts early. The final project was to be undertaken under a PhD student by helping further a certain point in their research. Topics are floated in class and one can make significant contributions.

Study Material and Resources

Lectures are the most important study material. Neuroscience by Dale Purves is referred to during lectures for the biology parts of the content.