397 datasets found

The dataset used in the paper is a collection of images of objects in 3D space, with multiple views of each object.

Robomimic environment consists of tasks such as lift, can, square, tool-hang, and transport.

D4RL benchmark suite consists of tasks such as locomotion, antmaze, adroit, and kitchen.

Offline reinforcement learning (RL) paradigm provides a general recipe to convert static behavior datasets into policies that can perform better than the policy that collected...

The dataset used in the paper is a collection of states sampled from a Markov Decision Process (MDP) using the random walk exploration method.

The dataset used in the paper is a collection of states sampled from a Markov Decision Process (MDP) using the RND exploration method.

The dataset used in the paper is a collection of states sampled from a Markov Decision Process (MDP) using the SMM exploration method.

The dataset used in the paper is a collection of states sampled from a Markov Decision Process (MDP) using the ChronoGEM exploration method.

The dataset used for the Gridworld tasks, consisting of 10K episodes of the agent following a uniform random policy.

The dataset used in the paper is the OpenAI Gym dataset, which consists of a set of environments for reinforcement learning.

The dataset used in the paper is the Arcade Learning Environment (ALE) dataset, which consists of 57 classic Atari games.

Interaction Networks: Using a Reinforcement Learner to train other Machine Learning algorithms

The Playground environment is a continuous 2D world. In each episode, N = 3 objects are uniformly sampled from a set of 32 different object types (e.g. dog, cactus, sofa, water,...

To overcome the sim-to-real gap in reinforcement learning (RL), learned policies must maintain robustness against environmental uncertainties. While robust RL has been widely...

The dataset used in the paper is a set of real-world robot learning tasks, including Create-Reacher, UR5-VisualReacher, Vector-ChargerDetector, and Franka-VisualReacher.

The 2D Grid World environment with various items and consequences

The Cartpole-v1 environment is used to evaluate the performance of Federated Reinforcement Distillation (FRD) framework.

MuZero, a model-based reinforcement learning algorithm that uses a value equivalent dynamics model.

The Google Research Football environment is a reinforcement learning experimental platform focused on training agents to play football.

The dataset is used for training a neural network consisting of 2 blocks of convolutional layers followed by max pooling layers and an LSTM block.