Carlos Florensa Campo

Graduate Student Researcher

7th floor Sutardja Dai Hall
UC Berkeley
Berkeley, CA 94720
United States
Email Address: 

Biography

Carlos is a PhD student in the Robotics Learning Lab at UC Berkeley, under the supervision of Prof. Pieter Abbeel. His main interest is solving robotics tasks in variable environments, with minimum supervision. This yields challenging sparse reward problems for Reinforcement Learning. He believes policy hierarchy, few shot learning, and automatic curriculum generation are key to solve these tasks, scale up to real world scenarios, and empower robotic systems.

Publications

Reverse Curriculum for Reinforcement Learning.
Carlos Florensa, David Held, Markus Wulfmeier, Pieter Abbeel.
CoRL 2017 [webpage][pdf][arXiv]
 
Stochastic Neural Networks for Hierarchical Reinforcement Learning.
Carlos Florensa;,Yan Duan, Pieter Abbeel.
ICLR 2017 [webpage][pdf][arXiv][code]
 
Capacity planning with competitive decision-makers: Trilevel MILP formulation, degeneracy, and solution approaches.
Carlos Florensa, Pablo Garcia-Herreros, Pratik Misra, Erdem Arslan, Sanjay Mehta, Ignacio E. Grossmann.
European Journal of Operations Research, 2017 [pdf]
 
“The magic of light!” - An entertaining optics and photonics awareness program.
Carlos Florensa, Miriam Martí, S. Chaitanya Kumar, Silvia Carrasco.
Education and Training in Optics and Photonics, 2013 [pdf]
 
Preprints:
Automatic Goal Generation for Reinforcement Learning Agents.
David Held*, Xinyang Geng*, Carlos Florensa*, Pieter Abbeel.
2017, arXiv:1705.06366. [webpage][arXiv]