Luis Barba

Luis Barba is a Research Fellow in the Machine Learning and Optimization group at École Polytechnique Fédérale de Lausanne (EPFL), Switzerland, working under Professor Martin Jaggi. He completed his PhD through a cotutelle program between Carleton University, Ottawa and Université Libre de Bruxelles, Brussels, supervised by Professors Stefan Langerman, Jit Bose, Pat Morin and Vida Dujmović. Prior to that, he earned his master's degree at Universidad Nacional Autónoma de México (UNAM) under Professor Jorge Urrutia.

Dr. Barba's research spans computational geometry, algorithms, graph theory, and more recently, machine learning and optimization. His work addresses fundamental problems in geometric data structures, Voronoi diagrams, graph coloring, and distributed learning. He has made significant contributions to understanding time-space trade-offs in geometric algorithms and developing efficient methods for problems like geodesic Voronoi diagrams and dynamic graph coloring.

Dr. Barba's publication record demonstrates a clear evolution from theoretical computational geometry to practical applications in machine learning. Early in his career, he focused on fundamental geometric problems including linear-time algorithms for geodesic Voronoi diagrams and efficient convex hull computation in polygonal domains with obstacles. More recently, his work has shifted toward machine learning, where he has developed novel optimization techniques for distributed and federated learning settings, including implicit gradient alignment methods and multilayer lookahead approaches.

Dr. Barba has published extensively in top-tier conferences and journals including Symposium on Computational Geometry (SoCG), Canadian Conference on Computational Geometry (CCCG), Algorithmica, and Discrete and Computational Geometry. His collaborative work demonstrates strong connections across the computational geometry and algorithms communities, with frequent co-authorship with leading researchers in these fields.

Throughout his career, Dr. Barba has maintained a consistent focus on algorithmic efficiency and computational complexity, whether addressing theoretical geometric problems or practical machine learning challenges. His work exemplifies how deep theoretical insights can inform practical computational approaches across different domains of computer science.