https://github.com/amanpriyanshu/rl-guessinggame

Can models adjust to truths and deception or can they replicate herd mentality. Best strategy simulation for Guessing Game. Where player can pick a single chit from a bucket of chits, which contain two types of chits, say Red and Blue, now those players that choose the correct chit get reward.
https://github.com/amanpriyanshu/rl-guessinggame

Last synced: 7 months ago
JSON representation

Can models adjust to truths and deception or can they replicate herd mentality. Best strategy simulation for Guessing Game. Where player can pick a single chit from a bucket of chits, which contain two types of chits, say Red and Blue, now those players that choose the correct chit get reward.

• Host: GitHub
• URL: https://github.com/amanpriyanshu/rl-guessinggame
• Owner: AmanPriyanshu
• License: mit
• Created: 2022-09-15T07:22:27.000Z (about 3 years ago)
• Default Branch: main
• Last Pushed: 2022-09-15T09:03:01.000Z (about 3 years ago)
• Last Synced: 2025-02-08T17:23:38.121Z (9 months ago)
• Language: Python
• Size: 104 KB
• Stars: 0
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# RL-GuessingGame

Can models adjust to truths and deception or can they replicate herd mentality. Best strategy simulation for Guessing Game. Where player can pick a single chit from a bucket of chits, which contain two types of chits, say Red and Blue, now those players that choose the correct chit get reward.

## Qualitative Analysis over Experiments

The RLAgent is successfully distinguish that majority sampling is to be done only when a sufficient number of passes is over. And rather it is prone to saying its own chit if it is among the first few sampling members.

We can show this with the demo_image.

![Figure](/demo_results_choice_of_option.png)

Here, each of these sub-buckets contain at most 8 samples.