Alfredo Torrico - The Greedy Algorithm: An Effective and Flexible Tool in Submodular Optimization

preview_player
Показать описание

Alfredo Torrico - Polytechnique Montréal

The Greedy Algorithm: An Effective and Flexible Tool in Submodular Optimization

Abstract: Constrained submodular maximization has a wide range of applications in multiple fields such as computer science, operations research, economics and machine learning. Nemhauser, Wolsey and Fisher introduced the standard greedy algorithm in 1978 and since then it has played an important role to obtain provable guarantees for different variants of the problem. Moreover, researchers have focused not only on providing improved quality and efficiency guarantees, but also on adapting it to related submodular optimization models. In this talk, I will address two questions: (1) Why does it perform so well in practice despite its worst-case guarantee? (2) How do we adapt it to a robust model that looks to simultaneously optimize several objective functions?

This talk is based on joint works with M. Singh, S. Pokutta, N. Haghtalab, J. Naor and N. Anari.
  1. Discrete Optimization Talks
  2. optimization
  3. operations research
  4. algorithms
Рекомендации по теме
Alfredo Torrico - 0:24:36

Alfredo Torrico - The Greedy Algorithm: An Effective and Flexible Tool in Submodular Optimization

10.6 Continuous Greedy, 0:35:20

10.6 Continuous Greedy, Part I

The role of 0:48:55

The role of subset selection in matching markets, Alfredo Torrico, 23/02/2022

Improved Multi-Pass Streaming 0:23:53

Improved Multi-Pass Streaming Algorithms for Submodular Maximization with Matroid Constraints

Greedy algorithm for 0:02:04

Greedy algorithm for submodular optimzation

10.3 Submodular Functions, 0:30:00

10.3 Submodular Functions, Part III

5B 1 Revisiting 0:20:02

5B 1 Revisiting Modified Greedy Algorithm for Monotone Submodular Maximization with a Knapsack Cons

10.2 Submodular Functions, 0:23:34

10.2 Submodular Functions, Part II

Behind the scenes 0:02:10

Behind the scenes

Submodular Maximization via 0:12:22

Submodular Maximization via Taylor Series Approximation (SDM21)

5.1 - Greedy 0:32:04

5.1 - Greedy Algorithm

EAAMO'22: Capacity Planning 0:18:39

EAAMO'22: Capacity Planning in Stable Matching: An Application to School Choice

EE596B Lecture 4, 1:34:20

EE596B Lecture 4, Submodular Functions, Optimization, and Applications to Machine Learning

Submodular Combinatorial Information 0:13:03

Submodular Combinatorial Information Measures with Applications in Machine Learning

constructex torrico fábrica 0:02:47

constructex torrico fábrica de placas de yeso

Submodularity and Optimization 1:27:39

Submodularity and Optimization -- Jeff Bilmes (Part 3)

Sensor placement optimization 0:01:19

Sensor placement optimization - RViz view

Streaming Weak Submodularity: 0:03:37

Streaming Weak Submodularity: Interpreting Neural Networks on the Fly (NIPS Preview Video)

Greedy Algorithms and 0:31:19

Greedy Algorithms and Matroids. MA252, University of Warwick, Week 9, Lecture 3

Rethinking for Secure 0:31:40

Rethinking for Secure Computation: A Greedy Approach - Prof. Muthuramakrishnan Venkitasubramaniam

Optimization in Machine 1:46:43

Optimization in Machine Learning (Lecture 9):Submodular Maximization and Greedy

Deeparnab Chakrabarty: Provable 0:24:37

Deeparnab Chakrabarty: Provable Submodular Function Minimization via Fujishige Wolfe Algorithm

Session 7-4 Representation 0:18:52

Session 7-4 Representation of Transversal Matroids with Applications in Parameterized Complexity

Greg Henselman (4/12/17): 1:07:41

Greg Henselman (4/12/17): Matroids & Canonical Forms Theory and Applications