Unlocking Personalization: A Deep Dive into Modern Recommendation Algorithms

This conference talk explores the evolution of recommendation systems, showcasing cutting-edge techniques driving today's sophisticated recommender engines. Journey through foundational heuristics, collaborative filtering approaches, and the application of two-tower neural networks, with emphasis on deep learning for recommendations. Discover the transformative role of large language models (LLMs) and how transformers enable highly personalized suggestions. Work with real-world datasets to implement and compare various recommendation algorithms using popular libraries and packages. Gain hands-on experience with the entire recommendation system pipeline, from data preprocessing to model deployment. Designed for data scientists, machine learning engineers, and researchers, this comprehensive presentation by Sarang Gupta at DSC EUROPE 24 in Belgrade equips participants with knowledge and skills to create state-of-the-art recommendation systems across diverse applications.