Lecture 6 part 1: ADMM (basic definitions and properties)
This is Lecture 6- part 1 - of the KTH-EP3260 Fundamentals of Machine Learning over Networks (MLoNs), lectured by Euhanna Ghadimi. This lecture reviews the basics and recent advances of the alternating direction method of multipliers (ADMM) for large-scale machine learning problems. In particular, this lecture covers fundamentals of dual ascent, dual decomposition, proximal methods, augmented Lagrangian, ADMM, convergence results, and hyperparameter tuning. Slides are available at the course website: https://sites.google.com/view/mlons/c...

▶︎
Lecture 6 part 2: ADMM (hyperparameter optimization and applications)

▶︎
Distributed Optimization via Alternating Direction Method of Multipliers

▶︎
Lecture 1 part 1: Introduction

▶︎
Duality: Lagrangian and dual problem

▶︎
25. Stochastic Gradient Descent

▶︎
Turing Award Winner: Disagreeing with Google, Postgres, Future Problems | Mike Stonebraker

▶︎
Daniel Kuhn: Data-driven and Distributionally Robust Optimization and Applications -- Part 1/2

▶︎
Penalty Function and Augmented Lagrangian methods 2013

▶︎
Constrained Optimization: Intuition behind the Lagrangian

▶︎
Understanding Lagrange Multipliers Visually

▶︎
Jonathan Eckstein - The ADMM, Progressive Hedging, and Operator Splitting (and workshop welcome)

▶︎
2020 ECE641 - Lecture 22: Augmented Lagrangian for Constrained Optimization

▶︎
Lecture 21: Alternating direction method of multipliers (ADMM)

▶︎
Why AI Hasn't Cured Anything...Yet, According to Jennifer Doudna | The Circuit

▶︎
Gauss — Das Geheimnis des größten Mathematikers aller Zeiten

▶︎
The French Do Not Care About Work

▶︎
Billionaire's WARNING: I'm SELLING. The Crash Is Already Here!

▶︎
A very, very basic introduction into distributed optimization

▶︎
