Robust-ML Research Group
Our research group focuses on the development of robust machine learning models with applications in health and human behavior analysis. We work on multi-view learning and conformal methods as a means to build robust systems. Our key research topics are:
- Multi-view learning algorithms
- Adaptive Multi-user systems
- Uncertainty estimation
- Machine learning for mental health
- Behavior analysis with wearable sensors
People
|
Name: Enrique Garcia Ceja Role: Principal Investigator Research topics: E-mail: e.g.mx (at) ieee.org |
|
Name: Iñaki González Morales Role: B.S Researcher Research topics: I am currently working on a machine learning research project focused on using ensemble models to improve classification performance on complex multi-view healthcare data. My main areas of interest include data science, applied machine learning, and backend development. E-mail: inakiglzm (at) outlook (.com) GitHub: https://github.com/InakiGonzalez |
|
Name: Cristóbal Estrada Salinas Role: B.S Researcher Research topics: Undergraduate researcher specializing in machine and deep learning applications. Experienced in developing ML classifiers with demonstrable results and creating data-driven solutions for real-world problems. Interests: Machine learning; Deep learning; Data analysis E-mail: a01174432 (at) tec.mx LinkedIn: https://www.linkedin.com/in/cestradasal GitHub: https://github.com/LAK3SHORE |
|
Name: Ricardo Kaled Corona Romero Role: M.S Researcher Research topics: My research focuses on improving model robustness through multi-view ensemble learning and topological data analysis (TDA). E-mail: hello (at) kaledcorona.xyz Website: https://kaledcorona.xyz GitHub: https://github.com/kaledcorona |
Funded projects
Project name: COMCAIIAM - Multi-View ML to combat cybersecurity thrats (translated from Spanish).
Description: We develop multi-view ML algorithms to detect and combat cybersecurity threats.
Date: 2025-Present
PI: Enrique Garcia-Ceja
Project name: SMIEAE - Intelligent Stress and Anxiety Monitoring System for University Students (translated from Spanish).
Description: In this project we use ambient and wearable sensors in University classrooms to monitor stress and anxiety using ML.
Date: 2024-Present
PI: Enrique Garcia-Ceja
Software
polystack Python library: Multi‑view stacking for scikit‑learn, with fast single‑view fallback and typed API. Author: Ricardo Kaled Corona Romero. Github: https://github.com/kaledcorona/polystack
Latest publications
Corona-Romero, K., Garcia-Ceja, E., Mendoza-Montoya, O. (2025). MNIST Dataset Classification via Multiview Topological Data Analysis. Mexican International Conference on Artificial Intelligence (MICAI).
Garcia-Ceja, E., Stautland, A., Riegler, M. A., Halvorsen, P., Hinojosa, S., Ochoa-Ruiz, G., … & Jakobsen, P. (2025). OBF-Psychiatric, a motor activity dataset of patients diagnosed with major depression, schizophrenia, and ADHD. Scientific Data, 12(1), 32.
Garcia-Ceja, E., Garcia-Banuelos, L., & Jourdan, N. (2025). Conformal prediction in multi-user settings: an evaluation. User Modeling and User-Adapted Interaction, 35(1), 5.
Mena-Martinez, A., Alvarado-Uribe, J., Garcia-Ceja, E., Barrera-Animas A., Escamilla-Ambrosio P. (2025). Benchmark of Wrist-Wearable Devices for Student Stress Monitoring, 17 Artificial Intelligence Mexican Congress.