Algorithmic Intelligence logo
Skip to content empty

Last updated on 2020-08-20Events > Past events

Past events 

Past events (2011).

AI Talk: 18/04/11 @ 13:00 in 2A08
Presenter: Tobias Mahlmann
Title: Towards Procedural Strategy Game Generation: Evolving Complementary Unit Types
Abstract: The Strategy Game Description Game Language (SGDL) is intended to become a complete description of all aspects of strategy games, including rules, parameters, scenarios, maps, and unit types. One of the main envisioned uses of SGDL, in combination with an evolutionary algorithm and appropriate fitness functions, is to allow the generation of complete new strategy games or variations of old ones. This presentation presents a first version of SGDL, capable of describing unit types and their properties, together with plans for how it will be extended to other sub-domains of strategy games. As a proof of the viability of the idea and implementation, an experiment is presented where unit types are evolved so as to generate complementary properties. A fitness function based on Monte Carlo simulation of gameplay is devised to test complementarity.

AI Talk: 11/04/11 @ 13:00 in 2A08
Presenter: Alejandro Arbelaez (Microsoft Research/INRIA)
Title: Learning During Search
Abstract: Autonomous Search is a new emerging area in Constraint Programming, motivated by the demonstrated importance of the application of Machine Learning techniques to the Algorithm Selection Problem, and with potential applications ranging from planning and configuring to scheduling. This area aims at developing automatic tools to improve the performance of search algorithms to solve combinatorial problems, e.g., selecting the best parameter settings for a constraint solver to solve a particular problem instance. In this talk, we study three different points of view to automatically solve combinatorial problems; in particular Constraint Satisfaction, Constraint Optimization, and SAT problems.

AI Talk: 28/03/11 @ 13:00 in 2A08
Presenter: Mark J. Nelson
Title: Technologies for interactive narrative
Abstract: Since the early 1980s, game designers and players have argued for more complex and interactive stories in games. Instead of linear stories or stories with a few branching decision points, the hope is that the player can exercise meaningful choices, resulting in a wide range of possible plots and experiences. Since it's infeasible for the game programmer to manually program what should happen in every possible situation once stories become more complex, some sort of technological support seems necessary. This talk will survey some of the approaches, including the ways stories and interaction in them can be represented, what kinds of criteria an AI story manager should be optimizing for (and how those are specified), and some current work at ITU on dynamically generating new parts of stories.

AI Talk: 14/03/11 @ 13:00 in 2A08
Presenter: Corrado Grappiolo
Title: Towards Multimodal Player Adaptivity in a Serious Game
Abstract: We present an initial demonstrator towards the creation of an adaptive serious game for teaching conflict reso- lution. The game demonstrator is built around a simple resource management conflict scenario. The game monitors data about the behaviour and the cognitive focus of the player and generates content driven by the modelled player state automatically in order to guide her toward particular learning targets.

Invited Talk: 7/03/11 @ 13:00 in 2A08
Presenter: Per Kristian Lehre
Title: Time-complexity Analysis of Evolutionary Algorithms
Abstract: Evolutionary algorithms (EAs) and other nature-inspired search heuristics have been applied successfully in many optimisation domains. However, a theoretical foundation that allows the performance of these heuristics to be predicted has until now been missing.

This talk will cover recent advances in time-complexity analysis of EAs. As a starting-point, we will estimate the expected optimisation time of the simple (1+1) EA on example problems. We will then consider techniques that allow more realistic EAs with complicated population dynamics to be handled. Finally, we will discuss black-box complexity theory which describes common limits on the efficiency of broad classes of randomised search heuristics.

AI Talk: 28/02/11 @ 13:00 in 2A08
Presenter: Noor Shaker
Title: Player-adaptive Procedural Content Generation
Abstract: Procedural content generation is an important technique for computer game development and has been used for both offline and online content generation. Emotions and player experience are critical in game design and they are usually triggered by game events. In this talk I will give an overview of our approach for online game adaptation that maximizes the player’s fun value in platform games.

AI Talk: 21/02/11 @ 13:00 in 2A08

AI Talk: 14/02/11 @ 13:00 in 2A08
Presenter: Philippe Bonnet
Abstract: ITU and the Smart Grid.

AI Talk: 07/02/11 @ 13:00 in 2A08
Presenter: Dario Pacino
Abstract: Dario Pacino will give an overview of his research during his semester abroad at Brown University.


Find this page Online