no code implementations • 28 Mar 2024 • Dimitris Bertsimas, Vassilis Digalakis Jr, Yu Ma, Phevos Paschalidis
We consider the task of retraining machine learning (ML) models when new batches of data become available.
1 code implementation • 22 Nov 2023 • Vassilis Digalakis Jr, Christos Ziakas
We present BackboneLearn: an open-source software package and framework for scaling mixed-integer optimization (MIO) problems with indicator variables to high-dimensional problems.
no code implementations • 20 Jul 2023 • Vassilis Digalakis Jr
Our work includes a detailed theoretical analysis of the distributed, differentially private entropy estimator which we use in one of our algorithms, as well as a detailed experimental evaluation, using both synthetic and real-world data.
no code implementations • 26 May 2023 • Dimitris Bertsimas, Vassilis Digalakis Jr
Owing to their inherently interpretable structure, decision trees are commonly used in applications where interpretability is essential.
1 code implementation • 22 Feb 2021 • Dimitris Bertsimas, Vassilis Digalakis Jr, Michael Linghzi Li, Omar Skali Lami
The algorithm is highly scalable, allowing us to train models with thousands of parameters.
no code implementations • 17 Jul 2020 • Dimitris Bertsimas, Vassilis Digalakis Jr
We present a novel approach for the problem of frequency estimation in data streams that is based on optimization and machine learning.
no code implementations • 11 Jun 2020 • Dimitris Bertsimas, Vassilis Digalakis Jr
We present the backbone method, a generic framework that enables sparse and interpretable supervised machine learning methods to scale to ultra-high dimensional problems.