A LUPI distillation-based approach: Application to predicting Proximal Junctional Kyphosis

Yun Chao Lin, Andrea Clark-Sevilla, Rohith Ravindranath, Fthimnir Hassan, Justin Reyes, Joseph Lombardi, Lawrence G. Lenke, Ansaf Salleb-Aouissi
Proceedings of the 9th Machine Learning for Healthcare Conference, PMLR 252, 2024.

Abstract

We propose a learning algorithm called XGBoost+, a modified version of the extreme gradient boosting algorithm (XGBoost). The new algorithm utilizes privileged information (PI), data collected after inference time. XGBoost+ incorporates PI into a distillation framework for XGBoost. We also evaluate our proposed method on a real-world clinical dataset about Proximal Junctional Kyphosis (PJK). Our approach outperforms vanilla XGBoost, SVM, and SVM+ on various datasets. Our approach showcases the advantage of using privileged information to improve the performance of machine learning models in healthcare, where data after inference time can be leveraged to build better models.

Cite this Paper

BibTeX
@InProceedings{pmlr-v252-lin24a, title = {A {LUPI} distillation-based approach: Application to predicting Proximal Junctional Kyphosis}, author = {Lin, Yun Chao and Clark-Sevilla, Andrea and Ravindranath, Rohith and Hassan, Fthimnir and Reyes, Justin and Lombardi, Joseph and Lenke, Lawrence G. and Salleb-Aouissi, Ansaf}, booktitle = {Proceedings of the 9th Machine Learning for Healthcare Conference}, year = {2024}, editor = {Deshpande, Kaivalya and Fiterau, Madalina and Joshi, Shalmali and Lipton, Zachary and Ranganath, Rajesh and Urteaga, Iñigo}, volume = {252}, series = {Proceedings of Machine Learning Research}, month = {16--17 Aug}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v252/main/assets/lin24a/lin24a.pdf}, url = {https://proceedings.mlr.press/v252/lin24a.html}, abstract = {We propose a learning algorithm called XGBoost+, a modified version of the extreme gradient boosting algorithm (XGBoost). The new algorithm utilizes privileged information (PI), data collected after inference time. XGBoost+ incorporates PI into a distillation framework for XGBoost. We also evaluate our proposed method on a real-world clinical dataset about Proximal Junctional Kyphosis (PJK). Our approach outperforms vanilla XGBoost, SVM, and SVM+ on various datasets. Our approach showcases the advantage of using privileged information to improve the performance of machine learning models in healthcare, where data after inference time can be leveraged to build better models.} }
Endnote
%0 Conference Paper %T A LUPI distillation-based approach: Application to predicting Proximal Junctional Kyphosis %A Yun Chao Lin %A Andrea Clark-Sevilla %A Rohith Ravindranath %A Fthimnir Hassan %A Justin Reyes %A Joseph Lombardi %A Lawrence G. Lenke %A Ansaf Salleb-Aouissi %B Proceedings of the 9th Machine Learning for Healthcare Conference %C Proceedings of Machine Learning Research %D 2024 %E Kaivalya Deshpande %E Madalina Fiterau %E Shalmali Joshi %E Zachary Lipton %E Rajesh Ranganath %E Iñigo Urteaga %F pmlr-v252-lin24a %I PMLR %U https://proceedings.mlr.press/v252/lin24a.html %V 252 %X We propose a learning algorithm called XGBoost+, a modified version of the extreme gradient boosting algorithm (XGBoost). The new algorithm utilizes privileged information (PI), data collected after inference time. XGBoost+ incorporates PI into a distillation framework for XGBoost. We also evaluate our proposed method on a real-world clinical dataset about Proximal Junctional Kyphosis (PJK). Our approach outperforms vanilla XGBoost, SVM, and SVM+ on various datasets. Our approach showcases the advantage of using privileged information to improve the performance of machine learning models in healthcare, where data after inference time can be leveraged to build better models.
APA
Lin, Y.C., Clark-Sevilla, A., Ravindranath, R., Hassan, F., Reyes, J., Lombardi, J., Lenke, L.G. & Salleb-Aouissi, A.. (2024). A LUPI distillation-based approach: Application to predicting Proximal Junctional Kyphosis. Proceedings of the 9th Machine Learning for Healthcare Conference, in Proceedings of Machine Learning Research 252 Available from https://proceedings.mlr.press/v252/lin24a.html.

Related Material