Completed
Geometry: Kernel Ratio Interval (KRI)
Class Central Classrooms beta
YouTube videos curated by Class Central.
Classroom Contents
Revisiting Nearest Neighbors from a Sparse Signal Approximation View
Automatically move to the next video in the Classroom when playback concludes
- 1 Intro
- 2 Revisiting Nearest Neighbors from a Sparse Signal approximation view
- 3 What is a neighborhood?
- 4 Neighborhood definitions: Kernels (Similarity)
- 5 Neighborhood definitions: Local linearity
- 6 Interlude: Sparse Signal Approximation
- 7 Neighborhood = Sparse signal approximation
- 8 Alternative: Basis pursuit
- 9 Neighborhoods: Non-negative basis pursuit
- 10 Non-Negative Kernel regression (NNK)
- 11 Geometry: Kernel Ratio Interval (KRI)
- 12 Example
- 13 Label propagation using graphs
- 14 Experiments: Label propagation
- 15 Experiments: Classification
- 16 Neighborhoods Summary
- 17 Conventional Approach
- 18 Solution: NNK-Means
- 19 kMeans vs Dictionary learning
- 20 Case study: Detecting unseen data using NNK-Means (representational outliers)
- 21 NNK-Means atom use in each scenario
- 22 NNK-Means for Outlier Detection
- 23 NNK-Means Summary
- 24 What is deep learning?
- 25 Graph based view of deep learning
- 26 NNK interpolation at penultimate layer
