2 papers
cs.LG2025
Adaptive Optimization for Prediction with Missing Data
Dimitris Bertsimas, Arthur Delarue, Jean Pauphilet
When training predictive models on data with missing entries, the most widely used and versatile approach is a pipeline technique where we first impute missing entries and then com…
cs.LG2025
Efficient Domain Adaptation of Multimodal Embeddings using Constrastive Learning
Georgios Margaritis, Periklis Petridis, Dimitris J. Bertsimas
Recent advancements in machine learning (ML), natural language processing (NLP), and foundational models have shown promise for real-life applications in critical, albeit compute-c…