Human decision-making data in the cart-pole game

Citation Author(s):
Varun
Dutt
Applied Cognitive Science Laboratory, Indian Institute of Technology Mandi
Mahavir
Dabas
Applied Cognitive Science Laboratory, Indian Institute of Technology Mandi
Shashank
Uttrani
Applied Cognitive Science Laboratory, Indian Institute of Technology Mandi
Aadhar
Gupta
Applied Cognitive Science Laboratory, Indian Institute of Technology Mandi
Submitted by:
Aadhar Gupta
Last updated:
Mon, 07/08/2024 - 15:58
DOI:
10.21227/zet8-tx79
Data Format:
Research Article Link:
License:
0
0 ratings - Please login to submit your rating.

Abstract 

The instantaneous state (situation) of the game was constituted by four values: the cart position, the cart speed, the pole angle to the vertical axis, and the pole angular velocity.

For each action taken by the human player in the game, a tuple containing the four values representing the current game situation, along with the action and reward obtained (utility), is recorded as a situation-decision-utility (SDU) tuple.

3 types of actions have been recorded: Move left (-1), move rght (1) and no action (0).

The cart-pole position and velocity are expressed in terms of pixels whereas the pole angle and angular velocity are expressed in terms of radians.

The cart-pole game: The source code for the cart-pole game used to collect the data is available at https://drive.google.com/drive/folders/1ug69x2vz3Gb4EYhjIv6gMiRZlZ9hQOzz....