-
FetchPush, FetchPickAndPlace and HandManipulateBlock
The FetchPush, FetchPickAndPlace and HandManipulateBlock environments from OpenAI gym robotics suite -
Dense Reward for Free in RLHF
The dataset used in the paper is not explicitly described, but it is mentioned that it is a preference dataset for language models. -
SAI Dataset
The dataset used for training the SAI agent, containing 7x7 Go games with multiple komi values. -
MuJoCo environments
The dataset used in the paper is not explicitly described, but it is mentioned that the authors used MuJoCo environments from the OpenAI gym. -
OpenAI Gym benchmark
The dataset used in the paper is the OpenAI Gym benchmark, which provides a set of environments for reinforcement learning. -
Funnel board
The Funnel board task is a domain where a ball falls through a grid of obstacles onto one of five platforms. Every other row of obstacles consists of funnel-shaped objects,... -
Room runner
The Room runner task is a domain where an agent moves through a randomly generated map of rooms, which are observed in 2D from above. The agent follows the policy of always... -
Discovering Blind Spots in Reinforcement Learning
The dataset used in the paper is a collection of oracle feedback, which is used to learn a blind spot model of the target world. -
Event Camera-based Reinforcement Learning
The dataset used in the paper is a simulated environment for event camera-based reinforcement learning. The dataset includes a car-like robot equipped with an event camera, and... -
Alchemy: A structured task distribution for meta-reinforcement learning
The Alchemy benchmark is a meta-learning environment rich enough to contain interesting abstractions, yet simple enough to make ne-grained analysis tractable. -
Mujoco control tasks
The authors used the Mujoco control tasks, including Ant-v2, HalfCheetah-v2, Hopper-v2, and Walker2d-v2. -
DCG-MAP-Elites-AI
The dataset used in this paper is a set of seven continuous control locomotion tasks implemented in Brax, derived from standard RL benchmarks. -
Pong and Breakout
The dataset used in the paper is a collection of tasks from OpenAI Gym, specifically Pong and Breakout. -
Sym-Q: Adaptive Symbolic Regression via Sequential Decision-Making
Symbolic regression holds great potential for uncovering underlying mathematical and physical relationships from empirical data. The authors introduce Symbolic Q-network... -
Multiple-confounded-Mujoco-Envs
The dataset used in the paper is a collection of environments with multiple confounders, including mass, length, damping, and a crippled leg. The dataset is used to evaluate the... -
HalfCheetah-v3
The dataset used in the paper is a modified version of the HalfCheetah-v3 environment from the OpenAI gym. -
FrozenLake-v0
The dataset used in the paper is a modified version of the FrozenLake-v0 environment from the OpenAI gym. -
PPO Cartpole Task
PPO Cartpole Task