Inspiration

Social cycling and conditional responses in the Rock-Paper-Scissors game, an academic research paper from Zhejiang University and the China State Key Laboratory of Theoretical Physics Link => https://arxiv.org/pdf/1404.5199v1.pdf

And Dom Dabish, for the inspiration and concept of an AI vs AI study.

What it does

A study into comparing the success rate of two competing AI and their approaches to predicting a rock-paper-scissors player's next move.

How I built it

16 hours of logic and research, 5 hours of programming.

Challenges I ran into

Logical planning for building two AI models simultaneously, as well as running them against each other in our local test environment. Also, using Firebase for discrete Markov chain memory.

Accomplishments that I'm proud of

Taking the time to learn time series analysis and prediction algorithms such as discrete Markov chains, hidden Markov models, etc. and learning how to implement an intelligent system with a clean UI using no third-party code.

What I learned

How to use discrete Markov chains for machine learning and temporal pattern recognition.

What's next for Janken

Mo' training, mo' winnin, mo' money. Training against as many humans as possible to

Share this project:

Updates