Gustavo Penha

I am a researcher in the field of information retrieval, interested in machine learning, ranking models, recommender systems, and natural language processing. Currently I work as a PhD under the supervision of Claudia Hauff at TU Delft. I had the opportunity of doing research at Amazon and Spotify during research internships. Other than science I am passionate about photography and I am working on my first photobook ๐Ÿ–ผ๏ธ.

news

Aug 26, 2022 Finished internship at Spotify research ๐ŸŽต.
Apr 19, 2022 We won the best paper award with our ECIRโ€™22 paper ๐Ÿ† ๐ŸŽ‰.
Aug 27, 2021 Finished my research internship at Amazon ๐Ÿ“ฆ: CHIโ€™22 & CHIIRโ€™22.

research interests

Representation learning for ranking
Text encoders learn representations for queries and documents, which are then used to calculate a relevance score. The goal is that relevant documents get close to the query and non-relevant documents get far from the query in the embedding space. I am interested in many aspects of representation learning, including negative sampling, disentanglement and interpretability.
Explainability and model understanding
Information filtering systems, such as document rankers and recommender systems, have a large impact into what we are able to find, what we are exposed to and the decisions we make. Understanding the behavior of such models, when they fail, how robust they are, and why they are recommending certain items over others is crucial for both machine learning practitioners and end users.

selected publications

  1. CHIIR short paper
    Pairwise Review-Based Explanations for Voice Product Search
    Penha, Gustavo, Krikon, Eyal, and Murdock, Vanessa
  2. ECIR ๐Ÿ† best paper
    Evaluating the Robustness of Retrieval Pipelines with Query Variation Generators
    Penha, Gustavo, Cรขmara, Arthur, and Hauff, Claudia
  3. RecSys
    What does BERT know about books, movies and music? Probing BERT for Conversational Recommendation
    Penha, Gustavo, and Hauff, Claudia
  4. RecSys ๐Ÿ† best paper RU
    Exploiting Performance Estimates for Augmenting Recommendation Ensembles
    Penha, Gustavo, and Santos, Rodrygo

presentations

Slides for our ECIR 2022 paper on query variations. video ๐ŸŽฌ.
Slides for the Glasgow IR seminar on 10 May 2021: video ๐ŸŽฌ.

academic services

  • Organizer at the Delft eXplainable AI Summer School 2022 (XAISS).
  • Organizer at Search-Oriented Conversatinal AI workshop (SCAI) at SIGIR'22.
  • Reviewer for ECIR (19, 20, 21, 22), CIKM (19, 21, 22), SIGIR (21, 22), RecSys (21, 22), WWWW'20, MICROS'21, CHIIR'22 and CUI'22.

one page cv