Carmelo Sferrazza

GaussGym: An open-source real-to-sim framework for learning locomotion from pixels

Flow Policy Gradients for Legged Robots

Bigger, Regularized, Categorical: High-Capacity Value Functions are Efficient Multi-Task Learners

ActSafe: Active Exploration with Safety Constraints for Reinforcement Learning

MaxInfoRL: Boosting exploration in reinforcement learning through information gain maximization

SOMBRL: Scalable and Optimistic Model-Based RL