Statements (47)
| Predicate | Object |
|---|---|
| gptkbp:instance_of |
gptkb:stadium
|
| gptkbp:bfsLayer |
3
|
| gptkbp:bfsParent |
gptkb:stadium
|
| gptkbp:analyzes |
Graphical interface in Open AI Gym
|
| gptkbp:based_on |
Classic control problem
|
| gptkbp:challenges |
Deep learning models
Beginner RL agents |
| gptkbp:depicts |
Graphs and plots
|
| gptkbp:designed_by |
gptkb:Open_AI
|
| gptkbp:difficulty |
gptkb:tank
|
| gptkbp:emulation |
Continuous time simulation
|
| gptkbp:has_gameplay_element |
Discrete action space with 3 actions
|
| gptkbp:has_goal |
Reach the flag at the top of the hill
|
| gptkbp:has_method |
Physics-based dynamics
|
| gptkbp:has_variants |
Different versions with modified parameters
|
| https://www.w3.org/2000/01/rdf-schema#label |
Mountain Car-v0
|
| gptkbp:is_analyzed_in |
Data scientists
|
| gptkbp:is_available_in |
Open AI Gym library
|
| gptkbp:is_compatible_with |
Various RL libraries
|
| gptkbp:is_documented_in |
Open AI Gym documentation
|
| gptkbp:is_evaluated_by |
Performance metrics like average reward
Reward-based evaluation |
| gptkbp:is_explored_in |
Machine learning practitioners
Reinforcement Learning research |
| gptkbp:is_implemented_in |
gptkb:Library
Gym library Numpy and Matplotlib |
| gptkbp:is_observed_in |
Continuous observation space with position and velocity
|
| gptkbp:is_part_of |
gptkb:software_framework
gptkb:Open_AI_Gym_Classic_Control Open AI Gym's collection of environments Reinforcement learning benchmarks |
| gptkbp:is_popular_in |
Academic settings
|
| gptkbp:is_similar_to |
gptkb:Cart_Pole-v0
|
| gptkbp:is_tested_for |
Researchers and developers
Other environments |
| gptkbp:is_used_for |
Algorithm comparison
Testing control algorithms |
| gptkbp:is_used_in |
Educational purposes
Benchmarking RL algorithms |
| gptkbp:number_of_episodes |
Maximum of 200 steps
|
| gptkbp:prize_money |
Reward of -1 for each time step until goal is reached
|
| gptkbp:requires |
Agent to learn to control a car
|
| gptkbp:state |
Car starts at a random position
Position and velocity of the car Vector representation of state |
| gptkbp:training |
Learning curve varies by algorithm
|