Course

Unit 1

Overview of Machine Learning and Deep Learning Models – Algorithm to hardware translation – Bitwidth – Fixed Point and Floating Point representations – Precision Effects

Unit 2

Least Mean Square Algorithm – Case Studies – Neural Network Implementations – Trade-offs

Unit 3

Advanced algorithms – Deep Learning implementations – Neuromorphic Architectures – Sparsity – Irregular Computations – Introduction to neuromorphic Architectures.

Course Objectives

• To introduce new paradigms in computing.
• To familiarize various aspects and issues in implementation of machine learning systems.
• To impart background on application of FPGAs and unconventional computing platforms for machine learning.
• To provide exposure to using state of the art computing tools.

Course Outcomes: At the end of the course, the student should be able to

• CO1: Ability to understand high performance machine learning architectures.
• CO2: Ability to apply computing paradigms for machine intelligence problems.
• CO3: Ability to suggest solutions and platforms for dataflow intensive problems.
• CO4: Ability to evaluate the use of diverse technologies to design efficient applications.

CO-PO Mapping:

Skills Acquired: Ability to develop architectures for Machine Learning.

1. Shiho Kim and Ganesh Chandra Deka, Advances in Computers, Volume 122: Hardware Accelerator Systems for Artificial Intelligence and Machine Learning, ScienceDirect, 2021
2. Andres Rodriguez, Deep Learning Systems: Algorithms, Compilers, and Processors for Large-Scale Production.Synthesis Lectures on Computer Architecture. Morgan & Claypool Publishers. Oct. 2020.
3. Sze, Designing Hardware for Machine Learning, in IEEE Solid-State Circuits Magazine, vol. 9, no. 4, pp. 46-54, Fall 2017 (paper)
4. Lei Deng, Guoqi Li, Song Han, Luping Shi, and Yuan Xie, Model compression and hardware acceleration for neural networks: A comprehensive survey. Proceedings of the IEEE 108, no. 4 (2020): 485-532. (paper)
5. Shail Dave, Riyadh Baghdadi, Tony Nowatzki, Sasikanth Avancha, Aviral Shrivastava, and Baoxin Li., Hardware acceleration of sparse and irregular tensor computations of ML models: A survey and insights. Proceedings of the IEEE 109, no. 10 (2021): 1706-1752. (paper)