About Me

Welcome! I am Felice Antonio Merra, a Senior Data Science Manager at Cognism. I am also serving as Industry Co-Chair at RecSys 2026.

Before joining Cognism, I worked for more than 3 years as an Applied Scientist at Amazon.com, contributing to Generative AI products (Rufus and AmazonQ), Information Retrieval (Amazon Search), Recommender Systems, and Natural Language Processing.

I hold a Ph.D. cum laude in Adversarial Machine Learning in Recommender Systems from the Department of Electrical Engineering and Information Technology, Polytechnic University of Bari — @PolibaOfficial — supervised by Prof. Tommaso Di Noia.

During my Ph.D., I completed a Summer Internship as an Applied Scientist at Amazon.com (Search and Personalization Team) and a research visit at the Knowledge Media Institute under the supervision of Prof. Enrico Motta.

Email: merrafelice@gmail.com

News & Awards

• [2026] [Service] Serving as Industry Co-Chair at RecSys 2026
• [2025] Best Short Paper — Do LLMs Memorize Recommendation Datasets? A Preliminary Study on MovieLens-1M — SIGIR 2025
• [2025] [ICML Paper] Hyperband-based Bayesian Optimization for Black-box Prompt Selection
• [2021] Best Short Paper — Runner Up — A Formal Analysis of Recommendation Quality of Adversarially-trained Recommenders — CIKM 2021
• [2021] Best Paper Award — MIT-IBM Watson AI Lab — Understanding the Effects of Adversarial Personalized Ranking Optimization Method on Recommendation Quality — AdvML@KDD 2021

Research Interest

My research activities mainly focus on artificial intelligence (AI). My investigation is devoted to novel approaches and applying machine learning (ML) algorithms, particularly to Trustworthy AI and Generative AI. In particular, I devote my attention to recommender system (RS) applications to study the robustness of modern ML recommender models affected by adversarial threats.

After having assessed the state-of-the-art of AML techniques in RS, I am investigating three main areas of study:

• the robustness of recommender models when affected by hand-engineered shilling attacks,
• the formal study of the effects of AML training strategies on the beyond-accuracy effects of recommenders, i.e., bias disparity, fairness, novelty
• the proposal of adversarial attacks against multimedia retrieval models.
• the Trustworthiness of ML (security, explainability, privacy, fairness, bias)

I continue to investigate AML approaches across ML tasks — including computer vision and reinforcement learning — with the goal of keeping end users at the core of my research and assessing how much they can trust an ML system.

For more information you can look at my Curriculum Vitae.

Publications

See my full list of publications on Google Scholar.

---