Robust Bichromatic Classification Using Two Lines
This collection of images and descriptions illustrates the concept of robust bichromatic classification using two lines to separate points in linear programming. Various scenarios, such as minimizing red/blue outliers and addressing violations, are explored along with different approaches for separating red and blue categories with and without outliers. The images provide visual representations of the mathematical concepts and algorithms involved in this classification method.
Download Presentation
Please find below an Image/Link to download the presentation.
The content on the website is provided AS IS for your information and personal use only. It may not be sold, licensed, or shared on other websites without obtaining consent from the author. Download presentation by click this link. If you encounter any issues during the download, it is possible that the publisher has removed the file from their server.
E N D
Presentation Transcript
Robust Bichromatic Classification using Two Lines Erwin Glazenburg, Thijs van der Horst, Tom Peters, Bettina Speckmann, Frank Staals 1
Robust Bichromatic Classification using Two Lines Separating points with lines Classification Linear Programming 2
Robust Bichromatic Classification using Two Lines Separate R and B, outliers - - - Perfect: Minimize red/blue outliers: Minimize both outliers: O(n) O(n log n) O((n + k^2) log n) (Chan 2005) 3
Robust Bichromatic Classification using Two Lines Separate R and B, many violations? 4
Robust Bichromatic Classification using Two Lines Separate R and B, many violations? 5
Separate R and B, two lines - All cases: O(n log n) (Hurtado et al, 2001, 2004) 6
3 Separate R and B: wedge, red outliers 2 1 LE 2 UE 35
Separate R and B: wedge, red outliers LE O(n log n) UE 36
Separate R and B, two lines with violations Thanks for listening :) 38
Vragen? 39
S3A: Robust Bichromatic Classification using Two Lines If you like: Envelopes 2023 1 line 2024 2 lines!! Duality Wedges and strips This thing: 40 Erwin Glazenburg, Thijs van der Horst, Tom Peters, Bettina Speckmann, Frank Staals