Unveiling the Evolution of AI Bots in Rock Paper Scissors

Table of Contents:

1. Introduction
2. The Rock Paper Scissors Game
3. The Story of Larry: an AI Bot
4. Larry's Knowledge Storage System
5. Larry's Prediction Process
6. Larry's Performance Against Different Bots
7. Larry's Evolutionary Capabilities
8. Larry's Weaknesses and Challenges
9. Conclusion

The Rock Paper Scissors Tournament: Larry and the Evolution of AI Bots

Introduction

In the world of skill comparison and competition, the easiest way to determine skill level amongst a group is to compete head to head. This can be done using either a tournament bracket format or a league table format. One such competition, known as the International Rule Shambo Tournament, is a rock paper scissors composition that ranks participants based on their match or game results against other participants. The interesting twist here is that the competitors in this tournament are artificial intelligence bots.

The Rock Paper Scissors Game

Before diving into the details of Larry, it's essential to understand the simplicity of the game itself. Rock paper scissors involves three possible inputs: rock, paper, and scissors. Each input has the same potential outputs, meaning that each input can win against one input and lose against another. This simplicity sets the stage for Larry's development and strategies.

The Story of Larry: an AI Bot

Larry, as we'll call it since it was not given an official name by its creators, was developed by Sony E. Valdes Vitruvius, John D. Barayga, and Procedural L. Fernandez as an AI bot to demonstrate the effectiveness of their theory against competitors in the first International Rule Shambo Tournament. Larry's main objective is to predict its opponent's moves based on their history and play the appropriate countermove.

Larry's Knowledge Storage System

To achieve its prediction capabilities, Larry utilizes a storage system based on arrays. These arrays represent the knowledge of its opponent's previous moves. One array stores the opponent's move history, while another array stores the most recent sequence of moves, known as the "window." The size of this window, denoted as 'n', needs to be defined before running the bot.

In the initial move, Larry always plays rock to avoid using a random number generator. After each round, Larry takes the opponent's last move and stores it in a variable called 'sec' (short for sequence). Larry then searches the move history for 'sec'. If the sequence is found, Larry notes the move that comes after it and adds it to a tally of rocks, Papers, and scissors.

This process is repeated for the entire move history, searching for other occurrences of the sequence. If the sequence is not found or the size of 'n' is larger than the history, the sequence size is decremented, and the process is repeated. Eventually, a sequence is found, and Larry chooses the highest tally as the prediction. In case of a tie, Larry counts the occurrences of the tied options across the entire history and uses the option with the highest count as the prediction. In the event of multiple ties, Larry employs a random number generator to select among the tied results and make the final prediction.

Larry's Prediction Process

Once Larry has made its prediction, it strategically plays the winning move against the opponent's most likely move. This predictive approach gives Larry an advantage against bots that do not have clear Patterns, such as the dummy bot that uses the value of Pi to determine its moves. As pi does not have any discernible patterns, Larry cannot predict its moves effectively. However, when faced with other history-based bots, Larry's performance varies.

Larry's Performance Against Different Bots

Larry struggles against two of the top-ranked history-based bots: the Direct History and the Iokan bots. The Direct History bot uses direct historical data to make its moves, while the Iokan bot employs complex decision-making strategies. In fact, the Iokan bot is so sophisticated that it can detect if a bot has predicted its algorithm and start to play against itself. Larry belongs to the group of bots that attempt to predict the Iokan algorithm, which poses a significant challenge.

On the other HAND, Larry has proven to be successful against bots that intentionally use patterns in their play, such as the "Switch a Lot" bot that never plays the same move twice and the "Go to a Rock" bot that only plays rock. These predictable patterns allow Larry to anticipate the opponent's moves and gain an upper hand in the game.

Larry's Evolutionary Capabilities

One remarkable characteristic of Larry is its evolutionary capabilities. During its development, the researchers did not explicitly code the behavior exhibited when facing the "Beat Freaking Pick" bot, which tallies the occurrences of rock, paper, and scissors in the past and uses the highest tally as its prediction. Larry, by attempting to predict its moves, demonstrated the ability to evolve and adapt beyond its programmed strategies.

Larry's Weaknesses and Challenges

While Larry has shown remarkable performance and adaptive qualities, it also has its weaknesses. Larry struggles against bots that have no discernible patterns and cannot effectively predict their moves. Additionally, the reliance on historical data limits Larry's ability to respond promptly to sudden changes or unpredictable strategies employed by opponents.

Conclusion

The Rock Paper Scissors Tournament provides an interesting platform for observing the capabilities of AI bots like Larry. Through its knowledge storage system and prediction process, Larry showcases an evolutionary approach to the game. While it faces challenges against highly complex bots, Larry demonstrates remarkable performance against pattern-based opponents. The ongoing evolution of AI bots in such competitions promises exciting advancements for the future.

FAQ:

Q: How does Larry predict its opponent's moves? A: Larry uses a knowledge storage system based on historical move data to predict its opponent's moves.

Q: What is the advantage of Larry's prediction process? A: Larry strategically plays the winning move against the opponent's most likely move, giving it an advantage in the game.

Q: Does Larry have any weaknesses? A: Yes, Larry struggles against bots that have no discernible patterns and cannot effectively predict their moves. Additionally, sudden changes or unpredictable strategies pose challenges for Larry's historical data-based approach.

Q: Can Larry evolve and adapt its strategies? A: Yes, Larry has demonstrated evolutionary capabilities by adapting its behavior beyond its initial programming when facing certain bots.

Resources:

• International Rule Shambo Tournament
• Direct History Bot
• Iokan Bot
• Beat Freaking Pick Bot