COMP 597 Fall 2021: Automated Reasoning with Machine Learning (4 credits)

Course Overview

Reasoning and learning are two fundamental piceces of intelligence (whether artificial or not). This course is designed to introduce the cutting-edge research on combining logical reasoning with machine learning. We will study the logical foundations and algorithms behind many reasoning engines, e.g., SAT solvers, SMT solvers, and domain-specific solvers. We will also study how machine learning can be used to improve these complicated reasoning systems. Specifically, this course will cover the following topics -- Boolean satisfiability (SAT), Satisifiability module Theories (SMT), program analysis, program synthesis, inductive logic programming, and neuro-symoblic methods.

Pre-requisites (recommended but not absolutely required):
  • MATH: 222 (Calculus), 223 (Linear Algebra), and 323 (Probability)
  • COMP: 251 (Algorithms), 302/303 (Programming), 424/451/551 (AI/Machine Learning)

Instructor

Xujie Si < email. >
Office: McConnell 324

Teaching Assistant

TBD

Lecture Schedule

Lectures: MW 11:35 AM -12:55 PM
Location: Wong Building 1050 (map)

The schedule below is tentative and subject to change throughout the semester.

Grading

Relevant Textbooks

The textbooks below are supplementary, since our main source of reading materials will be academic research papers.