معرفی
Prof. Noam Koenigstein is a Professor in the Department of Industrial Engineering at Tel Aviv University's Iby and Aladar Fleischman Faculty of Engineering. His research focuses on developing scalable machine learning solutions for real-world recommendation systems, with extensive industry collaborations including Microsoft (Xbox) and Yahoo! Music.
His primary research interests include recommender systems, collaborative filtering, and neural approaches to user modeling. He pioneered neural item embedding techniques like Item2Vec and advanced Bayesian methods for diversity-aware recommendations using Determinantal Point Processes. His work consistently bridges theoretical machine learning with industrial-scale deployment challenges.
Analysis of his publication record reveals a strong evolution from traditional matrix factorization toward deep learning and attention-based architectures, always emphasizing practical scalability and addressing cold-start problems. His research demonstrates exceptional continuity in solving core recommendation challenges across domains including e-commerce, music, and video services.
Scientific recognition includes:
- Best paper runner up award at TVX 2014 for group viewing pattern analysis
Prof. Koenigstein has led multiple industry-academia partnerships that produced deployable recommendation frameworks, notably for Xbox Movies and Windows Store. His invited RecSys 2017 talk highlighted critical gaps between academic research and industrial requirements in recommender systems, establishing him as a leading voice in practical recommendation science.
