Serdar Kadioglu Gives A Featured Talk On Multi-Objective Optimization At NVIDIA GTC
- Posted by Jesse Polhemus
- on March 17, 2022
Taking place online from March 21-24, 2022, NVIDIA GTC is a leading AI developer conference with more than 200,000 expected participants. This year, Brown CS Professor Serdar Kadioglu, also the Vice-President of Artificial Intelligence at Fidelity Investments, will be sharing recent work from his research team in a featured talk ("Multi-objective Optimization to Boost Exploration in Recommender Systems").
"Recommender systems," he explains, "are the backbone of personalized services that provide tailored experiences to individual users. Still, data sparsity remains a common challenge, especially for new applications where training data is limited or unavailable. We'll present a combinatorial optimization problem that formalizes the selection of item universe for experimentation in recommender systems. On one hand, a large set of items is desirable to increase diversity. On the other hand, a smaller set enables rapid experimentation and minimizes the time and the amount of data required to train machine learning models. We'll show how to optimize for such conflicting criteria using a multi-level optimization framework. Our approach integrates techniques from discrete optimization, unsupervised clustering, and latent text embeddings. Experimental results on well-known recommendation benchmarks demonstrate the benefits of optimized item selection."
For more information, click the link that follows to contact Brown CS Communication and Outreach Specialist Jesse C. Polhemus.
