Dragonfly: a modular deep reinforcement learning library

📅 2025-04-30
📈 Citations: 0
Influential: 0
📄 PDF
🤖 AI Summary
To address low algorithm reusability, high configuration maintenance overhead, and poor runtime efficiency of CPU-intensive simulation environments in deep reinforcement learning (DRL) experiments, this paper introduces a lightweight, modular open-source DRL library. Methodologically, it proposes a configuration-driven architecture based on JSON serialization, enabling full decoupling and zero-code switching of algorithmic components—including neural network architectures, samplers, and optimizers—and designs an efficient execution engine optimized for CPU-bound simulations via intelligent resource scheduling and batched inference. The library ships with standardized implementations of DQN and PPO, along with seamless integrations for Atari and MuJoCo benchmarks. Empirical evaluation demonstrates competitive or superior performance against RLlib and Stable-Baselines3 on standard benchmarks, while reducing configuration modification time by 90% and increasing simulation task throughput by 40%.

Technology Category

Application Category

📝 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.
Problem

Research questions and friction points this paper is trying to address.

Enables modular deep reinforcement learning experimentation
Facilitates parameter sweeps with minimal code maintenance
Optimized for CPU-intensive environments like simulations
Innovation

Methods, ideas, or system contributions that make the work stand out.

Modular deep reinforcement learning library
JSON serialization for swapping building blocks
Optimized for CPU-intensive numerical simulations
🔎 Similar Papers
No similar papers found.
J
J. Viquerat
MINES Paristech, CEMEF, PSL - Research University
P
Paul Garnier
MINES Paristech, CEMEF, PSL - Research University
A
Amirhossein Bateni
MINES Paristech, CEMEF, PSL - Research University
E
E. Hachem
MINES Paristech, CEMEF, PSL - Research University