Artificial Intelligence

CSCI-GA.2560
Monday 5:00-7:00
Warren Weaver Hall room 102.
Professor Ernest Davis

Reaching Me

Office hours: Tuesday 10:00-12:00, Wednesday 3:00-4:00, or by appointment.

Textbook:

Artificial Intelligence: A Modern Approach by Stuart Russell and Peter Norvig (3rd edition)

Prerequisites:

Fundamental algorithms.

Requirements:

Problem sets (collectively 30%), small programming assignments (20%), final exam (50%).

Description:

There are many cognitive tasks that people can do easily and almost unconsciously but that have proven extremely difficult to program on a computer. Artificial intelligence is the problem of developing computer systems that can carry out these tasks. This course will cover problem solving, automated reasoning, and machine learning.

Course topics:

Class notes

Propositional Logic
Davis-Putnam algorithm
Davis-Putnam: example.
Predicate calculus
Guide to expressing facts in first-order logic
Chapters 8 and 9 on probability and on random variables from my book Linear Algebra and Probability for Computer Science Applications are on the NYU Classes site. Read all of chap.8 and chap. 9 through section 9.3. There are a few errata, here.
Linear Separators and Support Vector Machines
K-means clustering
David Rosenberg's Slides on K-Means and EM maximization

Additional reading (Optional)

Commonsense Reasoning and Commonsense Knowledge in Artificial Intelligence, Ernest Davis and Gary Marcus, CACM, September 2015.
Article by Garry Kasparov about Computer Chess. (New York Review of Books, 2/11/2010)
The Master Algorithm by Pedro Domingos.
Gradient Descent -- Visualization
Joelle Pineau on Reinforcement Learning
David Silver's slides on AlphaGo (IJCAI16)

Schedule

Assignments

Problem set 1 Due Feb. 5 .
Programming Assignment 1 Due Feb. 12.
Problem set 2 Due Feb. 26.
Programming Assignment 2 Due Mar. 19
Problem set 3 Due Mar. 26
Programming Assignment 3 Due Apr. 9.
Problem set 4 Due Apr. 9.
Problem set 5 Due Apr. 23.
Programming Assignment 4 Due Apr. 30
    Test inputs and outputs

Submitting assignments

Problem sets are due at the beginning of class on the due date. I will accept them up to one week late with a penalty of 1 point out of 10. Problem sets should be submitted on the NYU Classes site.

Programming assignmemnts are due at the beginning of class on the due date. Except for the last assignment, these will be accepted up to two weeks late, with a penalty of 1 point out of 10 for each week late. Programming assignment should be submitted on the NYU Classes site; the form will be specified in each assignment.

Final Exam

The final exam will be given on Monday May. 14, 5:00-7:00, WWH 102. Outline of Final Exam
A sample exam and solutions is on the NYU Classes site.

Students with Disabilities

Academic accommodations are available for students with disabilities. Please contact the Moses Center for Students with Disabilities (212-998-4980 or mosescsd@nyu.edu) for further information. Students who are requesting academic accommodations are advised to reach out to the Moses Center as early as possible in the semester for assistance.

Cheating

You may discuss any of the assignments with your classmates (or anyone else) but all work for all assignments must be entirely your own. Any sharing or copying of assignments will be considered cheating. By the rules of the Graduate School of Arts and Science, I am required to report any incidents of cheating to the department. My policy is that the first incident of cheating will result in the student getting a grade of F for the course. The second incident, by GSAS rules, will result in expulsion from the University.