Catalog Description: Deep Networks have revolutionized computer vision, language technology, robotics and control. They have growing impact in many other areas of science and engineering. They do not however, follow a closed or compact set of theoretical principles. In Yann Lecun's words they require "an interplay between intuitive insights, theoretical modeling, practical implementations, empirical studies, and scientific analyses." This course attempts to cover that ground.
Units: 4.0
Student Learning Outcomes: Students will learn design principles and best practices: design motifs that work well in particular domains, structure optimization and parameter optimization., Understanding deep networks. Methods with formal guarantees: generative and adversarial models, tensor factorization., Students will come to understand visualizing deep networks. Exploring the training and use of deep networks with visualization tools.
Prerequisites: Math 53 and Math 54 or equivalent; Computer Science 70 or Statistics 134 or Electrical Engineering and Computer Sciences 126; Computer Science 61B or equivalent; Computer Science 189 (recommended)
Credit Restrictions: Students will receive no credit for COMPSCI L182 after completing COMPSCI 182, or COMPSCI W182. A deficient grade in COMPSCI L182 may be removed by taking COMPSCI 182, or COMPSCI W182.
Formats:
Fall: 3.0 hours of lecture and 1.0 hours of discussion per week
Spring: 3.0 hours of lecture and 1.0 hours of discussion per week
Grading basis: letter
Final exam status: Alternative method of final assessment