Aldo Pacchiano

About

Aldo Pacchiano is a leading researcher at the intersection of reinforcement learning (RL), evolutionary strategies (ES), and neural architecture search (NAS). His work focuses on making RL more scalable, robust, and practical for real-world applications, particularly in robotics and vision-based environments. Pacchiano’s major contributions include developing provably robust derivative-free optimization methods for RL, as demonstrated in his highly cited 2019 paper (11 citations), which showed that evolutionary strategies can match state-of-the-art policy optimization. He pioneered the ES-ENAS algorithm (2021, 9 citations), a scalable approach that combines ES with efficient NAS to automatically design compact RL policies without extra computational cost. His theoretical work on RL with once-per-episode feedback (2021, 3 citations) addresses a critical gap in real-world scenarios where rewards are sparse, while his Chromatic Networks (2019) and Implicit Attention (2021) papers tackle architecture search and visual overfitting, respectively. With over 27 citations across his top papers, Pacchiano’s research is shaping the future of data-efficient, robust RL systems.

Research Focus

Key Achievements

Top Papers

1. 1
   Provably Robust Blackbox Optimization for Reinforcement Learning

   11 citations · 2019
2. 2
   ES-ENAS: Combining Evolution Strategies with Neural Architecture Search at No Extra Cost for Reinforcement Learning.

   9 citations · 2021
3. 3
   On the Theory of Reinforcement Learning with Once-per-Episode Feedback

   3 citations · 2021
4. 4
   Reinforcement Learning with Chromatic Networks for Compact Architecture Search

   2 citations · 2019
5. 5
   Unlocking Pixels for Reinforcement Learning via Implicit Attention

   2 citations · 2021

Key Collaborators

Contact & Links