Esteban Real

Papers

Total Citations

H-Index

About

Esteban Real is a pioneering researcher at the intersection of artificial intelligence and automated machine learning, best known for his groundbreaking work on AutoML-Zero, which seeks to discover complete machine learning algorithms from scratch using only basic mathematical operations. His major contributions include the development of evolutionary search methods that can autonomously generate novel neural network architectures and learning algorithms, pushing the boundaries of what machines can invent without human intervention. Real’s most influential work, "Regularized Evolution for Image Classifier Architecture Search," has garnered over 1,200 citations, establishing him as a leading figure in neural architecture search. His 2023 paper, "Discovering Adaptable Symbolic Algorithms from Scratch," introduces AutoRobotics-Zero (ARZ), a method that creates zero-shot adaptable control policies for autonomous robots, enabling rapid adaptation to environmental changes without retraining. This work exemplifies his commitment to building AI systems that can invent their own solutions, with potential applications in robotics and beyond. Real’s research continues to inspire a new generation of scientists exploring the frontiers of automated discovery.