I am a Research Scientist at Soroco, working on foundation models for interaction data. Previously, I was a 2021 Computing Innovation Fellow and postdoctoral researcher jointly at MLD@CMU and Chicago Booth, where I worked with Bryon Aragam and Pradeep Ravikumar.
I received my PhD in Computer Science from Purdue University, where I was advised by Jean Honorio. Before my PhD, I completed my BSc in Mechatronics Engineering from the National University of Engineering in Lima, Peru.
I serve as Production Editor of the Journal of Machine Learning Research (JMLR), the flagship journal for the field of machine learning. For JMLR related inquiries please reach out to bello@jmlr.org.
Research
My research goal is to develop next-generation ML systems that will tackle some of the current major challenges, such as robustness, interpretability, and fairness. These systems necessitate a shift from standard statistical models that are susceptible to capture undesired nonlinear correlations to ones that can potentially discover (causal) relations from multimodal, complex data.
Topics of interest:
- Causal machine learning: Structure learning, invariant/causal representations
- Generative models: Probabilistic models, latent variable modeling
- Statistical learning: Structured prediction, sample complexity, exact inference
News
- 01/25: One paper accepted to AISTATS.
- 09/24: Two papers accepted to NeurIPS.
- 06/24: One paper accepted to NeuroImage.
- 04/24: One paper accepted to UAI.
- 09/23: Two papers accepted to NeurIPS.
- 09/23: Giving a talk on Oct. 17 about recent advances in continuous structure learning at the INFORMS Annual Meeting.
- 09/23: Presenting iSCAN at the Bay Area Machine Learning Symposium (BayLearn) on Oct. 19.
- 09/23: Code for iSCAN is now available on GitHub.
- 07/23: Talk about iSCAN at the Max Planck Institute for Intelligent Systems, Tübingen.
- 07/23: New preprint: iSCAN: New identifiability results of causal mechanism shifts among nonlinear SCMs without individual structure learning! (Accepted to NeurIPS).
- 06/23: New preprint: First set of results on global optimality for gradient-based DAG learning. (Accepted to NeurIPS).
- 06/23: Code for TOPO is now available on GitHub.
- 05/23: New preprint: TOPO: Optimization theory for continuous structure learning. (Accepted to ICML).
- 04/23: One paper accepted to ICML.
- 04/23: Awarded the DAAD AInet fellowship. A fund for visiting German institutions.
- 12/22: Code for DAGMA is now available on GitHub: Continuous optimization for structure learning with faster and more accurate log-det constraint.
- 09/22: Our paper “DAGMA: Learning DAGs via M-matrices and a Log-Determinant Acyclicity Characterization” has been accepted to NeurIPS.
- 09/22: Presenting DAGMA at the Bay Area Machine Learning Symposium (BayLearn) on Oct. 20!
- 05/22: One paper accepted to ISIT.
- 01/22: One paper accepted to AISTATS.
- 09/21: Started a joint postdoc at UChicago and CMU.
- 09/21: One paper accepted to NeurIPS.
- 06/21: Excited to have been awarded the Computing Innovation Fellowship!
- 04/21: One paper accepted to ISIT.
- 04/21: Talk at Pradeep Ravikumar’s lab at CMU on April 15.
- 04/21: Talk at David Sontag’s lab at MIT CSAIL on April 14.
- 04/21: Talk at Tomaso Poggio’s lab at MIT CBMM on April 5.
- 03/21: Honored to have been awarded the Bilsland Dissertation Fellowship at Purdue!
- 01/21: Talk at Peru’s 3rd Symposium of Deep Learning on Jan 29.
- 09/20: Talk at TECHSUYO’20 on October 29.
- 09/20: One paper accepted to NeurIPS.
- 05/20: Summer internship at Facebook AI.
- 01/20: One paper accepted to AISTATS.
- 09/19: One paper accepted to NeurIPS.
- 05/19: Summer internship at Facebook Ads Ranking team.
- 09/18: Two papers accepted to NeurIPS.
Service
Leadership
- Production Editor Journal of Machine Learning Research (JMLR)
- Mentor Data Science Institute Summer Lab, UChicago (2023)
- Web Chair LatinX AI Workshop at ICML 2020
Conference Reviews
Journal Reviews
- Journal of Machine Learning Research (JMLR)
- IEEE Trans. Pattern Analysis and Machine Intelligence (TPAMI)
- Transactions on Machine Learning Research (TMLR)
- Annals of Applied Statistics (AOAS)
- Journal of the Royal Statistical Society: Series B (JRSSSB)
- Journal of Computational and Graphical Statistics (JCGS)