This course introduces spectral graph theory along with applications to theoretical computer science, such as in error-correcting codes, analysis of sampling algorithms, and complexity theory.

Logistics

Instructor. Sidhanth Mohanty (Email me at sidhanth (dot) mohanty (at) northwestern (dot) edu)

When? 11:00 am to 12:20 pm, Tuesdays & Thursdays

Where? Tech LG62

Office Hours. 9:30 am to 11:00 am, Tuesdays & Thursdays, Mudd 3510; also by appointment

Evaluation. Homeworks and final project

Course discussion. Join the Piazza page.

Anonymous feedback. Please leave any feedback you have on the course and my teaching here.

Lecture notes

September 16

Introduction to spectral graph theory

September 18

The second eigenvalue

September 23

Vignette I: Expander codes

September 25

Spectral embeddings and Cheeger rounding

September 30

Vignette II: Spectral algorithms for planted problems

October 2

Alon–Boppana bound and Ramanujan graphs

October 7

Expander constructions I: Near-Ramanujan graphs via random graphs and lifts

October 9

Expander constructions II: Zig-Zag product

October 14

Interlacing families and Ramanujan graphs I

October 16

Interlacing families and Ramanujan graphs II

October 21

Sampling via Markov chains 101

October 23

High-dimensional expansion I: Local-to-global theorems

October 28

High-dimensional expansion II: Trickle-down

October 30

Vignette III: Rapid mixing of the basis exchange walk

Future topics (subject to significant revision)

November 4

Phase transitions in statistical physics

November 6

Vignette IV: Spectral condition in Ising models

November 11

Weak Poincaré inequalities

November 13

Slow-mixing random walks and locally stationary distributions

November 18

Spectral hypergraph theory: Kikuchi matrices and hypergraph Moore bound

November 20

The KLS conjecture: “every convex body is an expander”

November 25

Thanksgiving Break

November 27

Thanksgiving Break

December 2

Almost-resolution of KLS I: Eldan’s stochastic localization

December 4

Almost-resolution of KLS II

Homeworks

HW1

Due Date: October 28