Enric Boix

Enric Boix is an Assistant Professor of Statistics and Data Science at the University of Pennsylvania's Department of Statistics and Data Science. His research focuses on the theory of AI, including distillation, reasoning models, steering, monitoring, training dynamics, inductive bias of architectures, adversarial examples, feature learning, and AI safety. He teaches courses such as Foundations of Deep Learning and Seminars in Advanced Applications of Statistics. His work has been published in top-tier conferences and journals, including Science and NeurIPS.

Enric Boix-Adserà is an Assistant Professor at Wharton Statistics and Data Science, focusing on the theoretical foundations of machine learning and optimization. His research explores the interplay between algorithmic complexity, statistical learning, and the structure of deep neural networks. Boix-Adserà investigates how properties like the staircase property and merged-staircase property influence the learning dynamics of sparse functions and the convergence of stochastic gradient descent. His work also addresses computational challenges in optimal transport, multimarginal problems, and the complexity of learning models such as causal trees and transformers. By analyzing the average-case complexity of problems like clique counting and the hardness of multimarginal optimal transport, he contributes to understanding the limits and capabilities of modern machine learning algorithms.