Dust Presentation discussion about Artificial Intelligence engine

For a while now, I wanted to introduce you to a project that I started around two years ago. While not complete, the project has reached enough maturity so I can talk about it, show nice screen shots and envision its release.

This project is called , and it is a video game in the mindset of Build and Defend. Essentially, Dust contains many of the things that I wanted to add to B&D, but that I could not because it would have completely changed the game, or because it would have required to completely reprogram the game engine. Dust is a top-down, third person 2D game in a 3D world, where players try to survive in an post apocalyptic world filled with strange creatures as simple as that. The game is played real time (i.e. it is not turn based) in an gigantic and procedurally generated world populated with different areas and encounters. I would say that Dust is a Rogue Like (if you do not follow the very strict definition).

Continue reading “Dust Presentation discussion about Artificial Intelligence engine”

Machine Learning “for Dummies” (part 3) : The Decision tree and Decision Forest algorithms

In the last two posts (Machine Learning “for Dummies” [part 1] and Machine Learning “for Dummies” [part 2]) we introduced what was the “Classification” of Machine Learning. We also presented a very famous algorithm called “k-nearest neighbors”. If you have not read these posts, you may want to do it now to be sure to understand this post.

Continue reading “Machine Learning “for Dummies” (part 3) : The Decision tree and Decision Forest algorithms”

Machine Learning “for Dummies” (part 2) : The K-nearest neighbors algorithm

In the last post, we introduced what was the « Classification » of Machine Learning. We also presented some examples of applications (banks, hospitals, etc.) and we started to play with the Iris dataset. If you have not read the first post, now would be a good time in order to fully understand the following article.

Back to the Iris story: Remember, we have a list of iris flowers described by the length and width of their petals and sepals (the attributes). Also, we know the species of all except one iris (we will call this iris the “mysterious iris”), and we would like to find the species (or class) of the mysterious iris. For this task, we suppose the attributes of the irises to be indicative of the species.

Continue reading “Machine Learning “for Dummies” (part 2) : The K-nearest neighbors algorithm”

Machine Learning for Dummies (part 1) : Introduction to Classification

You might have already heard about the terms “Machine Learning” (or “Data Mining”, or “Big data”, or Data Analysis, or Data Science, or Cloud Computing) but you never actually knew what they meant. Maybe you know those are computer or mathematical stuffs and that is all. Maybe you have to use some softwares that relies on Machine Learning in your job, but for you, it is like magic.

Continue reading “Machine Learning for Dummies (part 1) : Introduction to Classification”

Creation and experiments with the Small Reverse Polish Notation Programming Language (Small for short)

1. Introduction

Do you know Lua? This is a simple scripting programming language with a very small interpreter (around 600kB once compiled) and that can be easily included and linked into any software. Essentially, Lua allows users to program and add new functions to an existing software without having to edit its source code and to recompile it. Lua is quite popular – especially in video game development: You would be surprised by the number of games that use Lua (World of Warcraft, Crysis, Civilization 5, SimCity 4, Stalker, etc. – https://en.wikipedia.org/wiki/Category:Lua-scripted_video_games for more)

Continue reading “Creation and experiments with the Small Reverse Polish Notation Programming Language (Small for short)”

Presentation and introduction

Hello stranger,

If you read this posts right after I publish it; if you read this post by exploring backward the hundreds of my posts; or if you find it on this old website that nobody updated for several years, I welcome you to my blog and to my first post.

My name is Mathieu Guillame-Bert. Today, I am working as a researcher in a university in the US. I am studding data mining for “large temporal databases”. I a few words, I am trying to make systems able to understand and predict the future. Some people (not me) call this “AI” (i.e. Artificial Intelligence). In my spare time, I make video games, I hack, I play, I read blogs, and I watch video on Youtube. It is not all I do, but you get the idea.

For a couple of years now, I had the desire to make a blog to talk about things I like. Today, I decided to do it. In this blog, I will talk about a lot of subjects: I will talk about my research and computer science in general in a very informal way. I will talk about making video games (e.g. artificial intelligence, mechanics, procedural generation, game design, experiments). I will talk about programming and hackings (e.g. creating my own minimal programming language, playing with c++, playing with phones, playing with the internet!). I will talk about little experiments and DIYs (e.g. my mini crossbow, my non-lethal mousetrap, my hydrogen balloons).

I have quite a precise idea of what will be my first couple of posts, and I have a large amount of subjects that I want to talk about but where I did not figured how to present them.

Nevertheless, I hope that you will enjoy reading this blog and I will see you in the next post.