IPBoost - Non-Convex Boosting via Integer Programming

By Marc E. Pfetsch, Sebastian Pokutta. 2020

6 citations recorded by the source corpus.

Recently non-convex optimization approaches for solving machine learning problems have gained significant attention. In this paper we explore non-convex boosting in classification by means of integer programming and demonstrate real-world practicability of the approach while circumventing shortcomings of convex boosting approaches. We report results that are comparable to or better than the current state-of-the-art.

Subfields: Artificial Intelligence, Computational Theory and Mathematics, Computer Networks and Communications

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Emilio Carrizosa
2. Dolores Romero Morales
3. Cristina Molero-Río
4. Birol Yüceoglu
5. S. Ilker Birbil
6. Mert Edali
7. Cynthia Rudin
8. Berk Ustun

In Computational Theory and Mathematics

1. Birol Yüceoglu
2. S. Ilker Birbil
3. Cristina Molero-Río
4. Dolores Romero Morales
5. Emilio Carrizosa
6. Mert Edali

In Computer Networks and Communications

1. Birol Yüceoglu
2. S. Ilker Birbil
3. Cristina Molero-Río
4. Dolores Romero Morales
5. Emilio Carrizosa
6. Mert Edali

Citations in the corpus

• Mathematical optimization in classification and regression trees (2021, methodology): “In some variants of Ada-Boost, the weights are optimized (Demiriz et al. 2002; Pfetsch and Pokutta 2020) with techniques such as column generation.”
• Rule Covering for Interpretation and Boosting (2020, methodology): “Another boosting method based on mathematical programming is called IPBoost [16].”