IEOR 8100 : Scheduling Algorithms Lecture 4

IEOR 8100-001: Learning and Optimization for Sequential Decision Making 04/06/16 Lecture 21: Learning and optimization for sequential decision making

IEOR 8100-001: Learning and Optimization for Sequential Decision Making

IEOR 8100: Topics in OR: Asymptotic Methods in Queueing Theory Fall 2009, Professor Whitt Class Lecture Notes: Wednesday, September 9. Heavy-Traffic Limits for the GI/G/1 Queue

We will start by considering this basic model, assuming unlimited waiting space and the first-come first-served service discipline (all customers served in order of arrival). This model is covered in almost all queueing textbooks. See Chapters III, VIII and X in Asmussen (2003), Applied Probability and Queues, second ed. A major treatise is Cohen (1982), The SingleServer Queue, second ed. Actua...

IEOR 8100-001: Learning and Optimization for Sequential Decision Making 02/03/16 Lecture 5: Thomposon Sampling (part II): Regret bounds proofs

We describe the main technical difficulties in the proof for TS algorithm as compared to the UCB algorithm. In UCB algorithm, the suboptimal arm 2 will be played at time t, if its UCB value is higher, i.e. if UCB2,t−1 > UCB1,t−1. If we have pulled arm 2 for some amount of times Ω( log(T ) ∆2 ), then with a high probability this will not happen. This is because after n2,t ≥ Ω(log(T )/∆), using c...

IEOR 8100 : “ Matchings ” Fall 2012 Lecture 21 : Distributed Greedy Maximum Weight Matching

Even though most distributed algorithms seem very simple at a first glance, the analysis used to prove their correctness and give bounds on the running time can be very challenging. To show the basic ideas used in analyzing distributed matching algorithms, an example of a simple greedy algorithm for the maximum weight matching will be considered. We will start by showing the approximation ratio...