معرفی
Marco Tulio Ribeiro is an Affiliate Assistant Professor at the University of Washington's Department of Computer Science & Engineering and a researcher at Google DeepMind. His work focuses on enhancing human interaction with machine learning models through explainability, debugging, and trust-building techniques. Ribeiro earned his Ph.D. from the University of Washington under advisors Carlos Guestrin and Sameer Singh.
His research interests span AI interpretability, model testing frameworks, and human-centered AI evaluation. Notable contributions include the CheckList tool for behavioral testing of NLP models (ACL 2020, Best Paper) and the Anchors framework for model-agnostic explanations (AAAI 2018). His work often bridges theoretical machine learning with practical human-AI collaboration challenges.
Ribeiro has been recognized with multiple awards, including the KDD 2016 Audience Appreciation Award and an ICML 2016 Best Paper Award for foundational interpretability work. His recent projects explore large language model reasoning (ART, 2023), interactive example curation (ScatterShot, 2023), and counterfactual explanations (Polyjuice, 2021).
His research philosophy emphasizes practical impact, as evidenced by his Medium blog posts on research project design and writing processes. While no formal student advisees are listed, his prolific co-authorships highlight collaboration with top researchers like Scott Lundberg, Tongshuang Wu, and Carlos Guestrin.
