ECE 591 607 Deep Learning Beyond Accuracy

3 Credit Hours

(Also offered as CSC 591 607)

In this course, students read and discuss research papers about deep neural neworks with a focus on not just accuracy but also resource consideration (e.g., FLOPs, parameter counts, time, memory, energy, size, etc.) With that interest, papers about techniques to design an efficient neural network architecture, such as structured/unstructured pruning, knowledge distillation, and quantization, will be read. On top of that, other dimensional metrics of machine learning, such as fairness, privacy, or sustainability, will also be explored. As a seminar course, this course is dedicated to paper reading, presentation, and discussion. Students will conduct a term project and take no exam. Students are expected to have implementation experiences on (convolutional deep) neural networks.

Course Objectives

To understand deep learning through the lens of resource (energy, time, size, memory, FLOPs, parameter count, etc) consumption.

Course Outline

The course is composed of paper reading, presentation, and discussion. And the students are expected to present their progress and results of a term project.

Course Requirements

Students are expected to have implementation experiences on convolutional deep neural networks.

Updated 6/21/2023