marek.k – Page 6

domain_randomization

Synthetic

Using domain randomization, we repeatedly randomize the simulation dynamics during training. No. of environments: 1

Dota 2 Matches

Games

Real World, Robotic

Explore player behavior and predict match outcomes.

driftgym

Synthetic

Gym environment for Car Drifting. No. of environments: 3

DROID

Agentic, Real World

DROID: A Large-Scale In-The-Wild Robot Manipulation Dataset

Drone Gym

Synthetic

Main purpose of this entire system is to investigate how human interaction can affect the traditional reinforcement learning framework No. of environments: 6

dsaa

Synthetic

Discrete State-Action Abstraction. No. of environments: 751

dtqn-gelu

Synthetic

Deep Transformer Q-Networks for Partially Observable Reinforcement Learning. No. of environments: 21

Dynamic pricing optimization

Multi, Pending

Robotic

Optimize pricing strategies dynamically in a competitive market environment

Ecobee Donate Your Data 1,000 homes in 2017

Energy

Agentic, Real World

anonymized ensemble of smart thermostat data and metadata from 1,000 U.S. single-family homes

EdNet is the dataset of all student-system interactions collected over 2 years by Santa, a multi-platform AI tutoring service with more than 780K users in Korea available through Android, iOS and web.

Author: marek.k