As a guest user you are not logged in or recognized by your IP address. You have
access to the Front Matter, Abstracts, Author Index, Subject Index and the full
text of Open Access publications.
In this paper, we introduce LibForce, a Deep Reinforcement Learning library built using C++ and LibTorch (PyTorch C++ Frontend). Deep reinforcement learning has reached great milestones and is expected to be implemented in the real world. We aim to realize the application of reinforcement learning in embedded devices. Embedded devices require short response times and small loads avoid runtime errors that Python programs may cause. Moreover, these devices have an affinity for C++ in software implementation. By supporting the implementation of reinforcement learning in C++, we can facilitate its implementation in embedded devices and support its deployment in the real world. Therefore, we developed LibForce, a reinforcement learning library in C++, especially for C++ beginners. A feature of LibForce is template parameters of a class are restricted with specified types using the C++ concept function to avoid C++ beginners’ misuse of classes. Moreover, the feature enables replacement of classes used in a reinforcement learning program. Herein, we present the specifications and implementation of LibForce, and show the effectiveness of the library with usage examples.
This website uses cookies
We use cookies to provide you with the best possible experience. They also allow us to analyze user behavior in order to constantly improve the website for you. Info about the privacy policy of IOS Press.
This website uses cookies
We use cookies to provide you with the best possible experience. They also allow us to analyze user behavior in order to constantly improve the website for you. Info about the privacy policy of IOS Press.