Organization of synaptic connectivity as the basis of neural computation and learning. Perceptrons, convolutional nets, and recurrent nets. Backpropagation and Hebbian learning. Engineering applications including computer vision and natural language processing.

Requirements

  • Two 80 minute lectures and one precept per week.
  • Grades (A-F) will be based on class participation (5%), problem sets (45%), midterm (20%), and final project (30%).
  • Participation includes speaking up in lecture and precept.
  • Participation also includes activity on Piazza—ideally asking good questions, giving good answers, and upvoting others’ contributions.

Prerequisites

  • Familiarity with linear algebra.
  • Basics of optimization and probability theory.
  • Knowledge of Python (or willingness to learn).

Lectures

Prof. Sebastian Seung will lecture twice a week.

Office Hours

Online discussions

You can ask and answer questions on the Piazza site. Piazza activity counts as class participation and can enhance your grade.

Final Project Guidelines

You will be expected to complete a mini-project on either a theoretical or an applied neural network problem. The project is worth 30% of your grade. You will collaborate in groups of 3-4 people with the help of our project mentors. More details about project ideas, milestones, and grade breakdowns can be found in the final project guidelines.

Homework assignments

Homework assignments will be due on Tuesdays, and should be submitted on gradescope.

Homework late policy

There is a accumulated percentage drop on the overall points of 20% per day (eg. 2 days late => -40%). If you have reasons (eg. conference, illness, …) for why you cannot make a deadline, write us (all instructors) an E-mail before the deadline including a suggested new deadline. We will grant requests on an individual basis.