publications

publications by categories in reversed chronological order. generated by jekyll-scholar.

2026

  1. gflownet_geometry.png
    Information-Geometric Forward Policy Training in GFlowNets
    Yordan Raykov, and Rodrigo Veiga
    Accepted to the NeurIPS 2026 Main Conference.
    arXiv
  2. diff_phdiag.jpg
    Denoising Score Matching with Random Features: Insights on Diffusion Models From Precise Learning Curves
    Anand Jerry George, Rodrigo Veiga, and Nicolas Macris
    In Proceedings of The 29th International Conference on Artificial Intelligence and Statistics, May 2026
    Spotlight. Republished in Journal of Statistical Mechanics: Theory and Experiment.
    arXiv HTML Journal

2025

  1. diff_manifold.jpg
    Analysis of Diffusion Models for Manifold Data
    Anand Jerry George, Rodrigo Veiga, and Nicolas Macris
    In 2025 IEEE International Symposium on Information Theory (ISIT), May 2025
    arXiv HTML PDF

2024

  1. sgf_fig04.jpg
    Stochastic Gradient Flow Dynamics of Test Risk and its Exact Solution for Weak Features
    Rodrigo Veiga, Anastasia Remizova, and Nicolas Macris
    In Proceedings of the 41st International Conference on Machine Learning, 21–27 jul 2024
    arXiv HTML PDF

2023

  1. ml_perc_red.jpg
    Learning curves for the multi-class teacher–student perceptron
    Elisabetta Cornacchia, Francesca Mignacco, Rodrigo Veiga, and 3 more authors
    Machine Learning: Science and Technology, Feb 2023
    arXiv HTML PDF

2022

  1. arXiv_fig01_image.jpg
    Phase diagram of Stochastic Gradient Descent in high-dimensional two-layer neural networks
    Rodrigo Veiga, Ludovic Stephan, Bruno Loureiro, and 2 more authors
    In Advances in Neural Information Processing Systems, Feb 2022
    arXiv HTML PDF
  2. rbm.jpg
    Restricted Boltzmann Machine Flows and The Critical Temperature of Ising models
    Rodrigo Veiga, and Renato Vicente
    Feb 2022
    Preprint
    arXiv PDF

2020

  1. ICU.jpg
    Age-structured estimation of COVID-19 ICU demand from low quality data
    Rodrigo Veiga, Rodrigo Murta, and Renato Vicente
    Feb 2020
    Preprint
    arXiv PDF