[Eecs_mscs] New Course Offering: EE 6900: Hardware for Deep Learning (Monday, Wednesday, Friday from 12:55 PM - 1:50 PM in ARC 101)

[Hunter, Tiffany]

New Course Offering:

EE 6900: Hardware for Deep Learning (Monday, Wednesday, Friday from 12:55 PM - 1:50 PM in ARC 101) 3 credit hours

This course is intended to provide graduate students with an in-depth study of the underlying hardware needed for deep learning and machine learning applications. As the neural network model size and complexity increases for improved accuracy, computational complexity and energy consumption increases proportionally. In this course, students will understand deep network computations for vision and image processing applications using hardware accelerators. Techniques to reduce the computation burden such as quantization, optimized dataflow and mapping, pruning and compression will also be discussed in this course. As data movement plays a crucial role in hardware mapping and optimizations, the design of interconnects for hardware accelerators for various neural network models such as CNN, LSTM, RNN, transformer and attention models will be discussed.

Thanks
Avinash.


------------------------------------------------------------------------

Avinash Karanth

Director & Chair, School of Electrical Engineering and Computer Science

Joseph K. Jachinowski Professor in EECS

Associate Editor - IEEE Transactions on Computers

Associate Editor - IEEE Transactions on Cloud Computing

Ohio University, Athens, OH 45701.

Phone: 740-597-1481

Webpage: https://oucsace.cs.ohio.edu/~avinashk
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.ohio.edu/pipermail/eecs_mscs/attachments/20231114/97d0afe3/attachment.html>