Completed
Non-backtracking for trust and centrality: avoid the echo chamber
Class Central Classrooms beta
YouTube videos curated by Class Central.
Classroom Contents
Physics-Inspired Algorithms and Phase Transitions in Community Detection - 2014 Symposium
Automatically move to the next video in the Classroom when playback concludes
- 1 Intro
- 2 What is structure?
- 3 Statistical inference
- 4 The stochastic block model
- 5 Assortative and disassortative
- 6 Likelihood and energy
- 7 Statistical significance
- 8 What's the best labeling?
- 9 Belief propagation (a.k.a. the cavity method)
- 10 The Karate Club: leaders vs. followers
- 11 The Karate Club: two factions
- 12 Two local optima in free energy
- 13 Active learning: update the model as we learn more
- 14 The double life of Belief Propagation
- 15 A phase transition: detectable to undetectable communities
- 16 Phase transitions in semisupervised learning
- 17 Hierarchical clustering
- 18 Clustering nodes with eigenvalues
- 19 When does this work?
- 20 The non-backtracking operator
- 21 Comparing with standard spectral methods
- 22 Non-backtracking for trust and centrality: avoid the echo chamber
- 23 Morals
- 24 Physics culture meets machine learning
- 25 Challenges
- 26 Shameless Plug
