Are you working in your MSc or PhD dissertation? Do you want to present your ongoing work to the AI4G school lecturers, and get some valuable insight on your work?
If so, drop us an e-mail with your name, affiliation, supervisor, your (tentative) dissertation title and a brief summary of your goals. We'll give you 5 to 10 minutes to present your work!
Note: Students from outside of IPCA that are pesenting their work on the Students Session can register with the lowest registration fee!
Procedural Content Generation 101
This class will introduce you to the magical world of procedural content generation (PCG), the class will consist in a review of the history of PCG, its usage in the industry and academia, and a summary of various helpful techniques. Time permitting we will do a small exercise in content creation.
AI Planning Questions
This course will begin with illustrating Game AI Planning in F.E.A.R.: playing the game and then looking at in-game planning data, what we can learn from it. Then, we will try to give answers to the most frequently asked questions about Game AI Planning, divided in 4 topics: References (what should I read? What is available? …), Usage (what is planning for in a game? …), Structures (how does Game AI planning works? What does it need to work? …), and Development (who takes Game AI Planning decisions? What are the coding constraints for Game AI Planning? …).
Following the illustration with F.E.A.R., we will compare in-game planning data from 6 FPS (F.E.A.R., Killzone 3, Transformers 3, Shadow of Mordor, Rise of the Tomb Raider and Shadow of the Tomb Raider) in order to show how planning works in these games, the differences between the kind of planning which occurs in these games and how Planning Analytics can help understand and improve Game AI Planning.
Music in games: can AI help developers? (yes)
This class will introduce you to how the algorithmic music field, with particular attention to how that relates to games. We will also discuss how AI can help the music designers in the gaming industry, what I think is holding us back, and how could we go forth (by telling you a bit about my research). You will also be introduced to some tools for music/audio generation, and hopefully you will be able to create your first generative audio system.
Evolving neural networks for playing video games
This class will introduce you to evolutionary algorithms, and specially: Genetic Algorithms. Then you will learn about Neural Networks and reinfocerment learning. Using an Evolutionary Algorithm, you will see how it is possible to evolve a Neural Network to pilot a racing car. Finally, you will see deveral applications of evolved neural networks on the gaming industry.
Voxel based Pathfinding with Jumping for Games
In this class, we will discuss the limitations of the Navigation Meshes when it comes to taking jumps into consideration. Then, a new solution using grid-based any-angle pathfinding is shown. In this context, the performance is critical and the paths should be optimal and efficient. Our results show that it is possible to successfully apply the proposed solution in game development.
AI Characters in Games
Rui Prada & Samuel Mascarenhas
In this class we will discuss the role of AI characters in games and present some concerns and good practices for creating believable and compelling characters. The FAtiMA toolkit (https://fatima-toolkit.eu/) will be presented. We will discuss the core fundamental theories behind its development and discuss its architecture and features, and discuss how it supports character creation.
FAtiMA Toolkit Hand-on
Rui Prada & Samuel Mascarenhas
In this class we will go through the FAtiMA toolkit usage scenarios and show how to create a conversation scenario with AI characters. Participants will get the chance to try the toolkit and develop their own scenarios.
Game Theory in large incomplete information games
In this presentation, we will discuss Game Theory in general and especially its applications in the resolution of problems with very deep decision trees, namely large sequential problems that can be represented as games, where the decision makers have to deal with stochastic variables. Most of these approaches will be exemplified with Poker which is a game that's being deeply used by the community to measure scientific advances in this domain.