Roberto Cipollone

DIAG, La Sapienza University of Rome, Italy.

photo1.jpg

I am a 3rd year PhD candidate in Engineering in Computer Science, at La Sapienza University of Rome. I am working in the WhiteMech research group with Giuseppe De Giacomo. In 2020, I graduated in Master of Science in Artificial Intelligence and Robotics, and I’ve previously completed a Bachelor of Computer and System Engineering at the same university.

My PhD research mainly focuses on Reinforcement Learning. Specifically:

  • learning in non-Markovian and history-dependent environments;
  • hierarchical Reinforcement Learning and state-actions abstractions.

My background also covers other areas of Machine Learning and Formal Methods.

I have been TA for both Master’s and Bachelor’s courses. I have been reviewer for many AI conferences. This small website is just a preview. Contact me at my personal email cipollone.rt@gmail.com for a full CV.

news

Mar 26, 2024 I’ll present my NeurIPS paper at the RL theory seminars. Don’t miss it live, or on YouTube afterwards!
Dec 10, 2023 I was at NeurIPS 2023 in New Orleans, USA!
Sep 30, 2023 I was at ECAI 2023 in Kraków, Poland.
Sep 14, 2023 I was at the EWRL 2023 workshop in Brussels, Belgium.

latest posts

selected papers

  1. Provably Efficient Offline Reinforcement Learning in Regular Decision Processes
    Roberto Cipollone, Anders Jonsson, Alessandro Ronca, and Mohammad Sadegh Talebi
    In Proceedings of Neural Information Processing Systems (NeurIPS), 2023
  2. Exploiting Multiple Abstractions in Episodic RL via Reward Shaping
    Roberto Cipollone, Giuseppe De Giacomo, Marco Favorito, Luca Iocchi, and Fabio Patrizi
    In Thirty-Seventh AAAI Conference on Artificial Intelligence (AAAI), 2023