RL-Glue: Language-Independent Software for Reinforcement-Learning Experiments

By Brian Tanner, Adam White. 2009

129 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Science Applications, Molecular Biology, Reinforcement Learning, Evaluation & Benchmarks

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Science Applications

In Molecular Biology

In Reinforcement Learning

In Evaluation & Benchmarks

Citations in the corpus

OpenAI Gym (2016, background): “…released, such as the Arcade Learning Environment (ALE) [5], which exposed a collection of Atari 2600 games as reinforcement learning problems, and recently the RLLab benchmark for continuous control [6], to which we refer the reader for a survey on other RL benchmarks, including [7, 8, 9, 10, 11].”
Dopamine: A Research Framework for Deep Reinforcement Learning (2018, background): “RLGlue (Tanner & White, 2009) emphasized benchmarking of value-based algorithms across a variety of canonical tasks, and drove a yearly competition; BURLAP (MacGlashan, 2016) provides object-oriented support; PyBrain (Schaul et al., 2010) supports neural networks, and is perhaps closest to more…”
Behaviour Suite for Reinforcement Learning (2019, background): “Further, bsuite facilitates clear, targeted and uniﬁed experiments across diﬀerent code frameworks, something that can help to remedy issues of reproducibility in RL research (Tanner & White, 2009; Henderson et al., 2017).”
Reinforcement learning in feedback control (2011, methodology): “In parallel, software environments such as RL-Bench (Tanner and White 2009) or CLSquare (Riedmiller et al. 2006) have been developed that support easy study and standardized benchmarking of tasks in many different domains.”
SkyAI: Highly modularized reinforcement learning library (2010, methodology): “We also have a plan to make an interface to the RLGlue [17] which is a language-independent software package for RL experiments.”
The Reinforcement Learning Competitions (2010, methodology): “" This event, as well as all later events, relied on RL-Glue (White 2006, Tanner and White 2009), a platformand language-independent interface for connecting agent, environment, and experiment programs .”
How Humans Teach Agents - A New Experimental Perspective (2012, methodology): “1 A screenshot of RL-Library Tetris actions.”
The Malmo Platform for Artificial Intelligence Experimentation (2016, background): “Con-ceptually, its abstraction is inspired by RLGLue [ Tanner and White, 2009 ] in that the high-level components are agents that interact with an environment by perceiving observations and rewards, and taking actions.”
A video game description language for model-based or interactive learning (2013, background): “The library’s availability as an open-source package, under the non-restrictive BSD license, is an invitation for others to use and build upon it, for example to diversify RL competitions [24], [25], more generally to benchmark learning algorithms in the games domain, or even to use it in teaching.”
RLLib: C++ Library to Predict, Control, and Represent Learnable Knowledge Using On/Off Policy Reinforcement Learning (2015, unknown): “Notably, the most common RL development platform is RL-GLUE(1) with RL-LIBRARY(2) packages [29].”
Deep Reinforcement Learning: An Overview (2017, methodology): “• RLGlue (Tanner and White, 2009) is a language-independent software for RL experiments.”
Value Function Approximation in Reinforcement Learning Using the Fourier Basis (2011, methodology): “* Sarah Osentoski is now with the Robert Bosch LLC Research and Technology Center in Palo Alto, CA. 1 Open-source, RL-Glue (Tanner and White, 2009) compatible Java source code for the Fourier basis can be downloaded from http://library.rl-community.org/ wiki/Sarsa_Lambda_Fourier_Basis_(Java). demons”
Empirical evaluation methods for multiobjective reinforcement learning algorithms (2011, methodology): “The server capability provided by RL-Glue provides a possible solution to this approach—agents can connect remotely to the server and execute the new policy within the environment.”
Neuroevolutionary reinforcement learning for generalized control of simulated helicopters (2011, methodology): “However, after the competition, the software, based on RL-Glue [85], was made public.”
Algorithms for Reinforcement Learning (2010, methodology): “It is a free, language-neutral software package that implements a standardized RL interface (Tanner and White, 2009).”