Basic Information

  • Course Code: EE 746
  • Course Name: Neuromorphic Engineering
  • Course Offered In: 2023-2024
  • Semester Season: Autumn
  • Instructors: Prof. Udayan Ganguly
  • Prerequisites: Some coding experience in MATLAB/Python.
  • Difficulty (1 being easy and 5 being tough): 2

Course Content

Different phenomena were studied in detail with focus on the mathematical understanding of how processes work:

  1. The biological neuron, how it works and transmits electrical signals. Action potential was explained followed by modelling of the neuron.
  2. Synapse and learning were taught, in specific, varying electrical potentials and the motion of ions across membranes.
  3. Networks of neurons and modern methods to model these. Boltzmann machines and liquid state machines were studied.

Feedback on Lectures

Highly recommend attending lectures. The lectures are based on discussion of concepts and active class participation. They go into the detail of how each process and phenomenon works. Slides are followed and are explained very well, in an easy to understand way. Highly recommend taking notes in class.

Feedback on Evaluations

There were 4 quizzes (15% each) (including midsem and endsem) and 2 programming assignments (Total 10%). Class participation had 5% weightage and the final project (Teams of 3) had 25% weightage.

Questions in the quizzes were directly from the discussions conducted in class. They were logic based and required an understanding of the basic concepts of how neurons function. The programming assignments involved simulation of neurons and their activity in python or MATLAB. The final project involved going through research papers, implementing them and adding some novelty. The final evaluation was through submission of code and a presentation. The project topics had a lot of variety, from AI topics to devices topics. Software required varied from topic to topic. Some projects used LTspice while others had python coding.

Study Material and Resources

Dale Purves- Neuroscience JG Taylor- Computational Neuroscience

Follow-up Courses

No follow up as such, but you can take up an RnD project in their lab if you are interested in exploring the subject further.

Final Takeaway

You will find exams very easy if you attend classes and participate in discussions. The project is a great way to get started on reading, understanding and implementing research papers, followed by some innovation.