Autor: Martin Haring

In the previous blog I thought through a Stakeholder map and the most important personas. Knowing who will primarily use the tool, it is now time to think through what their use cases should look like – it is time to map the user journey.

In this blog, I explain my approach of this tools user journey as well as a first mock-up of the tool.

In the previous blog I shared my first thoughts about the tool I aim to develop, including the most probable target groups and the biggest issues that may arise for them. To get a clearer image of the situation, I will use three design methods that will specify the target groups and their needs.

In this blog, I introduce tools – a stakeholder map, a persona and a project rundown.

I aim to create a tool that applies a Reinforcement Learning (RL) bot to any given Unity application. The bot shall then try to maximize a game’s built-in score, which will be the only requirement for the tool to work. Functionality-wise, I see no major obstacles in the tools developement – there are a few things to clear up design-wise though.

In this blog, I will go over some thoughts concerning the design of this (still) unnamed RL tool for Unity. This blog marks the beginning of (another) new blog series I will dedicate to the tools‘ design.

Typically, the algorithm that an Reinforcement Learning (RL) bot is following is built around the reward function. In the previous Lost Chapters blog I mentioned Curriculum Learning: a method that has the purpose of improving the algorithm – and there are more performance improvers like that, which are today’s topic.

In this blog, I explain methods that improve an RL AI’s algorithm.

I aim to create a tool that lets an AI play a Unity game. The first step is to look for existing projects in the field – which will be the topic of today’s blog.

In this blog, I will go over existing tools and frameworks for Unity and briefly explain some common testing methods.

I recently read a Master Thesis in the field of Reinforcement Learning – and realized that there is a lot of important theoratical content I skipped over so far. This blog marks the beginning of the „Lost Chapters“ series, which will cover topics I might have missed during my journey through the AI jungle.

In this blog, I will take a glance at the design of reward functions.

In diesem Blogeintrag analysiere ich die Masterarbeit ‚Using Reinforcement Learning for Games with Nondeterministic State Transitions‘ von Max Fischer, eingereicht 2019 an der Linköping University im Department of Computer and Information Science.

Over the past few weeks I developed a small roll-a-ball project and took some close looks into its compoments. I tested tools that could extract the projects assets or scripts. I managed to fetch the ingame score of the project from the Windows registry. Now its time to explain why – its time to introduce this topics‘ main objective.

In this (anniversary edition) blog, I present the big picture of „Reinforcement Learning in Unity“.

PlayerPrefs are stored in the registry. What that actually is and how its contents can be extracted is today’s topic.

In this blog, the Windows registry is introduced; followed by a small guide on how to extract some of its contents into a text file.

The hacking tools introduced in the last blog are very good for modding purposes. In my case, I want to extract pieces of information – in particular, the score – while the game is running. Since I developed the roll-a-ball game in question I have a clue what to look into at this point: PlayerPrefs.

In this blog, I introduce the PlayerPrefs concept.