Dragonfly: a modular deep reinforcement learning library

Jonathan Viquerat, Paul Garnier, Amir-Hossein Bateni, Elie Hachem

Abstract: Dragonfly is a deep reinforcement learning library focused on modularity, in order to ease experimentation and developments. It relies on a json serialization that allows to swap building blocks and perform parameter sweep, while minimizing code maintenance. Some of its features are specifically designed for CPU-intensive environments, such as numerical simulations. Its performance on standard agents using common benchmarks compares favorably with the literature.

External IDs:dblp:journals/corr/abs-2505-03778