Marcos M. Raimundo

Institute of Computing - University of Campinas

Google Scholar GitHub LinkedIn Lattes Twitter Email

Marcos M. Raimundo is an Assistant Professor at University of Campinas (UNICAMP), in the Institute of Computing (IC). He is also a faculty member of the REasoning for COmplex Data laboratory (Recod.ai). His research interests include Machine Learning, Multi-objective Optimization, Ethical AI, Mathematical Programming.

Interests

Machine Learning
Multi-objective Optimization
AI Ethics
Mathematical Programming

Education

Ph.D. in Computer Engineering, 2018 School of Electrical and Computer Engineering, University of Campinas

M.Sc. in Computer Engineering, 2014 School of Electrical and Computer Engineering, University of Campinas

B.Sc. in Computer Engineering, 2011 University of Campinas

Experience

Adjunct Professor — Institute of Computing, University of Campinas

Campinas · May 2022 – present

Teaching of Computing and Machine Learning for Computer Science and Engineering. Research and Graduate Students advising. Extension Programs. Administrative Roles.

Adjunct Professor — Federal University of Rio de Janeiro

Rio de Janeiro · Sep 2021 – May 2022

Teaching of Statistics and Operations Research for Production and Gas Engineering. Research and Graduate Students advising. Extension Programs. Administrative Roles.

Postdoctoral Researcher — School of Applied Mathematics, FGV

Rio de Janeiro · Oct 2019 – Aug 2021

Developed a project analyzing criminality in São Paulo and Rio de Janeiro using Machine Learning, Visualization and Optimization tools, including an interpretability tool that creates counterfactual scenarios to reduce criminality, built using combinatorial optimization and machine learning, plus a visualization tool for exploring counterfactual scenarios across multiple, similar regions.

Data Scientist — Tech - Inovação Tecnológica para a Agropecuária

Piracicaba · May 2019 – Sep 2019

Data Analysis. Optimization Modeling. Staff training in Machine Learning. Software Engineering.

Research Assistant — University of Campinas

Campinas · 2012 – Dec 2018

Operations research applied to inventory and vehicle routing management, multiobjective optimization, machine learning applied to detection of epileptic seizures and ant genera identification, and applications of multiobjective optimization in multi-class, multi-label, transfer, multi-task, and multi-view learning.

Publications

PCE: Model-Agnostic Plausible Pareto-Optimal Counterfactual Explanations

Arthur Hendricks Mendes de Oliveira, Giovani Valdrighi, Marcos Medeiros Raimundo (2026). arXiv preprint arXiv:2606.18418.

Balancing Performance and Inclusion: A Novel Reject Inference Framework for Credit Scoring

Athyrson Machado Ribeiro, Marcos M Raimundo (2026). The 2nd Workshop on Advances in Financial AI Workshop: Towards Agentic and Responsible Systems.

HybridMorph: Towards Usage of Synthetic with Real Data for Medical MR Images

Saumya Roy, Deepak Mishra¹, Marcos M Raimundo (2026). Computer Vision, Pattern Recognition, Image Processing, and Graphics: 10th National Conference, NCVPRIPG 2025, Jammu, India, July 16–18, 2025, Revised Selected Papers, Part I.

Long-term Fairness with Selective Labels

Giovani Valdrighi, Isabel Valera, Marcos M Raimundo (2026). Forty-third International Conference on Machine Learning.

Navigating the Rashomon Set: The Impact of Score Distributions and Decision Thresholds on Model Agreement

Giovani Valdrighi, Marcos M Raimundo (2026). Algorithmic Fairness Across Alignment Procedures and Agentic Systems.

Towards a Geometric Theory of Fairness: Detecting Mode Collapse on the Grassmannian Manifold

Beatriz Cardoso Nascimento, Marcos M Raimundo (2026). ICLR 2026 Workshop on Geometry-grounded Representation Learning and Generative Modeling.

When AI Describes Race? Unveiling Racial Bias in Vision-Language Models in Brazilian People

Leodécio Braz da Silva Segundo, Marcos M Raimundo (2026). Algorithmic Fairness Across Alignment Procedures and Agentic Systems.

Reasoning Distillation and Structural Alignment for Improved Code Generation

Amir Jalilifard, Anderson de Rezende Rocha, Marcos Medeiros Raimundo (2025). arXiv preprint arXiv:2510.17598.

Balancing Performance and Reject Inclusion: A Novel Confident Inlier Extrapolation Framework for Credit Scoring

Athyrson Machado Ribeiro, Marcos Medeiros Raimundo (2025). arXiv preprint arXiv:2510.12967.

Best practices for responsible machine learning in credit scoring

Giovani Valdrighi, Athyrson M. Ribeiro, Jansen SB Pereira, Vitoria Guardieiro, Arthur Hendricks, Décio Miranda Filho, Juan David Nieto Garcia, Felipe F. Bocca, Thalita B. Veronese, Lucas Wanner, Marcos Medeiros Raimundo (2025). Neural Computing and Applications.

CounterCrime-using counterfactual explanations to explore crime reduction scenarios

Marcos M Raimundo, Germain Garcia-Zanabria, Luis Gustavo Nonato, Jorge Poco (2025). IEEE Transactions on Visualization and Computer Graphics.

Combinatorial U-Curve Optimization for Multi-Task Learning: A Task Selection Approach

Giovana K Bonás, Marcelo S Reis, Marcos M Raimundo (2025). 2025 International Joint Conference on Neural Networks (IJCNN).

Multi-criteria rank-based aggregation for explainable AI

Sujoy Chatterjee, Everton Romanzini Colombo, Marcos Medeiros Raimundo (2025). 2025 International Joint Conference on Neural Networks (IJCNN).

M²FGB: A Min-Max Gradient Boosting Framework for Subgroup Fairness

Jansen Silva de Brito Pereira, Giovani Valdrighi, Marcos Medeiros Raimundo (2025). Proceedings of the 2025 ACM Conference on Fairness, Accountability, and Transparency.

Pareto sensitivity, most-changing sub-fronts, and knee solutions

Tommaso Giovannelli, Marcos Medeiros Raimundo, Luis Nunes Vicente (2025). arXiv preprint arXiv:2501.16993.

Evaluating Hate Speech Detection to Unseen Target Groups

Alexandre Negretti, Marcos M Raimundo (2024). Conferência Latino-Americana de Ética em Inteligência Artificial.

Mining Pareto-optimal counterfactual antecedents with a branch-and-bound model-agnostic algorithm

Marcos M Raimundo, Luis Gustavo Nonato, Jorge Poco (2024). Data Mining and Knowledge Discovery.

Exploring the Trade-off Between Model Performance and Explanation Plausibility of Text Classifiers Using Human Rationales

Lucas E Resck, Marcos M Raimundo, Jorge Poco (2024). Findings of the Association for Computational Linguistics: NAACL 2024.

Enforcing fairness using ensemble of diverse Pareto-optimal models

Vitória Guardieiro, Marcos M Raimundo, Jorge Poco (2023). Data Mining and Knowledge Discovery.

Analyzing the Equity of the Brazilian National High School Exam by Validating the Item Response Theory’s Invariance.

Vitoria Guardieiro, Marcos M Raimundo, Jorge Poco (2022). EDM.

Exploring multiobjective training in multiclass classification

Marcos M Raimundo, Thalita F Drumond, Alan Caio R Marques, Christiano Lyra, Anderson Rocha, Fernando J Von Zuben (2021). Neurocomputing.

CriPAV: street-level crime patterns analysis and visualization

Germain Garcia-Zanabria, Marcos Medeiros Raimundo, Jorge Poco, Marcelo Batista Nery, Cláudio T Silva, Sergio Adorno, Luis Gustavo Nonato (2021). Preprint.

Multi-criteria analysis involving Pareto-optimal misclassification tradeoffs on imbalanced datasets

Marcos M Raimundo, Fernando J Von Zuben (2020). 2020 International Joint Conference on Neural Networks (IJCNN).

An extension of the non-inferior set estimation algorithm for many objectives

Marcos M Raimundo, Paulo AV Ferreira, Fernando J Von Zuben (2020). European Journal of Operational Research.

Exploring multi-objective optimization to deal with multiple feature extractions in detection of epileptic seizures

M.M. Raimundo, F.J. and Von Zuben (2019). Journal of Epilepsy and Clinical Neurophysiology.

Investigating multiobjective methods in multitask classification

Marcos M Raimundo, Fernando J Von Zuben (2018). 2018 International Joint Conference on Neural Networks (IJCNN).

Ant genera identification using an ensemble of convolutional neural networks

Alan Caio R Marques, Marcos M Raimundo, Ellen Marianne B Cavalheiro, Luis FP Salles, Christiano Lyra, Fernando J Von Zuben (2018). PloS one.

Ensembles of Multiobjective-Based Classifiers for Detection of Epileptic Seizures

Fernando S Beserra, Marcos M Raimundo, Fernando J Von Zuben (2017). Iberoamerican Congress on Pattern Recognition.

Many-objective ensemble-based multilabel classification

Marcos M Raimundo, Fernando J Von Zuben (2017). Iberoamerican Congress on Pattern Recognition.

Multi-objective transfer learning for epileptic seizure detection

F.S. Beserra, M.M. Raimundo, F.J. and Von Zuben (2017). Journal of Epilepsy and Clinical Neurophysiology.

Um algoritmo adaptativo para gerar soluções Pareto-ótimas via branch-and-bound no espaço de objetivos

Marcos M. Raimundo, Fernando J. Von Zuben (2014). XLVI SIMPÓSIO BRASILEIRO DE PESQUISA OPERACIONAL.

Unveiling Hemodynamic Footprints: A Manifold Learning Approach for Systemic Disease Classification using ECG Signals

Joao Victor PB Avanzini, Marcos M Raimundo (YYYY). Preprint.

Projects

Long-Term Impacts of Machine Learning-Supported Decisions

A two-stage sequential decision-making framework combining offline reinforcement learning, contextual bandits, and multi-objective optimization to eliminate harmful feedback loops, mitigate selective labels (reject inference), and ensure long-term fairness and policy flexibility.

Multi-objective Optimization and Ensemble Learning for Machine Learning Models

Research on continuous and many-objective optimization techniques (NISE/MONISE) applied to regularized machine learning, model selection, ensemble generation, and algorithmic fairness.