skrl: Modular and Flexible Library for Reinforcement Learning
Authors: Antonio Serrano-Muñoz, Dimitrios Chrysostomou, Simon Bøgh, Nestor Arana-Arexolaleiba
JMLR 2023 | Venue PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Examples, in simulation and in the real world, of use cases with their respective scripts and description of functionalities are included as well as benchmarking results. |
| Researcher Affiliation | Academia | Antonio Serrano-Mu noz1 EMAIL Dimitrios Chrysostomou2 EMAIL Simon Bøgh2 EMAIL Nestor Arana-Arexolaleiba1,2 EMAIL 1 Department of Robotics and Automation, Mondragon Unibertsitatea, Arrasate, Spain 2 Department of Materials and Production, Aalborg University, Aalborg, Denmark |
| Pseudocode | No | The paper describes the implementation and features of the skrl library but does not include any structured pseudocode or algorithm blocks for its own methodology. |
| Open Source Code | Yes | The library s documentation can be found at https://skrl.readthedocs.io and its source code is available on Git Hub at https://github.com/Toni-SM/skrl. |
| Open Datasets | Yes | In addition to supporting environments that use the traditional interfaces from Open AI Gym / Farama Gymnasium, Deep Mind and others, it provides the facility to load, configure, and operate NVIDIA Isaac Gym, Isaac Orbit, and Omniverse Isaac Gym environments. |
| Dataset Splits | No | The paper describes an RL library that interacts with various environments but does not provide specific training/test/validation dataset splits, as is typical for traditional supervised learning tasks. |
| Hardware Specification | No | The paper mentions a "GPU-based physics simulation platform from NVIDIA" and "tens of thousands of simultaneous environments on a single GPU," but does not specify exact GPU models, CPU models, or other detailed hardware specifications used for experiments. |
| Software Dependencies | No | skrl is an open-source modular library for RL written in Python (on Py Torch (Paszke et al., 2019) and JAX (Bradbury et al., 2018)). The paper mentions PyTorch and JAX but does not provide specific version numbers for these or other software components. |
| Experiment Setup | No | The paper describes the architecture and features of the skrl library, including supported algorithms and components. It states that "Examples, in simulation and in the real world, of use cases with their respective scripts and description of functionalities are included as well as benchmarking results" in the documentation, but does not provide concrete experimental setup details, hyperparameters, or training configurations within the main text. |