Thesis Proposal – Developing a Risk-Aware Blackjack AI Using Distributed Reinforcement Learning

Background

Blackjack is one of the world’s most popular card games. Its simple rules, however, conceal a complex element of risk assessment and strategic decision-making. The objective is to get a hand sum as close to 21 as possible without exceeding it, while also beating the dealer’s hand. This balancing act between taking risks to win and avoiding “busting” (going over 21) makes the game an excellent case study for examining decision-making under uncertainty.

Project Description

This thesis project aims to construct a Blackjack-playing AI that uses Distributed Reinforcement Learning to develop its risk awareness. The project will explore how an AI, starting with only the basic rules of the game, progressively adapts its strategy and risk assessment over time by playing a large number of hands.

The goal is to evaluate the AI’s ability to learn an optimal strategy by analyzing its playing style from the first hand to the Xth hand. Special attention will be paid to how the AI handles situations where the risk of busting is high.

Implementation

The game simulation and AI development can be implemented in any programming language. A key part of the project is to visualize the results in a Graphical User Interface. The GUI should allow the user to configure game variables, such as the number of decks used, after which the AI will act based on these inputs.

For a successful thesis work, we believe that the applicants should have:

– Bachelor/ master studies in software engineering,  engineering physics or equivalent

– Interest in AI and programming.

– Good knowledge in both Swedish and English.

To give you the best possible support during your thesis, we’d like you to be able to come to the office connected to the project and spend most of your time working from there.

Application:

We look forward to receiving your resume, and preferably, a personal letter in which you explain why you want to write your thesis with Syntronic.

We screen and evaluate applications on an ongoing basis. The thesis project may be filled before the application deadline.