Explore advanced concepts in online linear programming and learning through this comprehensive lecture from the 2019 ADSI Summer Workshop on Algorithmic Foundations of Learning and Control. Delve into resource allocation, comparative ratios, and dynamic learning as Stanford University's Yinyu Ye presents "Further Developments on Online Linear Programming and Learning." Examine key ideas, impossibility results, and convergence theories while gaining insights into stochastic processes, competitive ratios, and the YG algorithm. Discover practical applications, closed-loop solutions, and simulation results that demonstrate the power of these techniques. Engage with a generic framework and explore the large L regime to enhance your understanding of cutting-edge algorithmic approaches in learning and control systems.
2019 ADSI Summer Workshop- Algorithmic Foundations of Learning and Control, Yinyu Ye
Paul G. Allen School via YouTube
Overview
Syllabus
Introduction
Summary
Resource Allocation
Toy Example
Linear Programming
Comparative Ratio
Impossible Result
Possible Result
Key Idea
Dynamic Learning
General Linear Programming
Competitive Ratio
YG
Convergence
Optimal Solution
Proof
Stochastic Process
Stop in Time
Theorems
Applications
Evaluation
Closed Loop Solution
Future
Closed Loop
Simulation
Results
Generic Framework
Large L Regime
Taught by
Paul G. Allen School
