Class Central is learner-supported. When you buy through links on our site, we may earn an affiliate commission.

YouTube

The Non-Stochastic Control Problem - Elad Hazan

Institute for Advanced Study via YouTube

Overview

Explore the cutting-edge developments in linear dynamical systems control during this 58-minute computer science seminar. Delve into the application of new machine learning methods that challenge classical control theory assumptions. Learn about efficient control techniques for systems with adversarial noise, general loss functions, unknown dynamics, and partial observation. Discover how these advancements are revolutionizing fields such as robotics, finance, engineering, and meteorology. Gain insights from speaker Elad Hazan of Princeton University as he presents findings from collaborative research efforts. Follow the seminar's progression through topics including optimal control, dynamics, loss functions with memory, and the concept of minimum regret in non-stochastic control problems.

Syllabus

Introduction
Control Theory
Optimal Control
Dynamics
Loss Function with Memory
The Holy Grail of Control
NonStochastic Control
Minimum Regret
Intuition
Representation Theorem
Interpretation
Summary

Taught by

Institute for Advanced Study

Reviews

Start your review of The Non-Stochastic Control Problem - Elad Hazan

Never Stop Learning.

Get personalized course recommendations, track subjects and courses with reminders, and more.

Someone learning on their laptop while sitting on the floor.