Faculty habilitation
Group : Learning and Optimization

Contributions to Evolutionary Design of Embodied Agents

Starts on 07/12/2009
Advisor :

Funding :
Affiliation : Université Paris-Saclay
Laboratory :

Defended on 07/12/2009, committee :

Research activities :
   - Artificial Intelligence
   - Evolutionary computation
   - Robotics
   - Evolutionary Algorithms

Abstract :
This Habilitation Thesis manuscript presents some of my research activities as Assistant Researcher ("Maître de Conférences") at Université Paris-Sud XI since 2003. It contains a selected summary of my contributions to the ?eld of Evolutionary Design of autonomous agents and developmental systems. Evolutionary Design is an automatic design method based on evolutionary operators, ie. combining selection pressure and partly blind variations, and has been shown to be particularly relevant to address problems for which a success value may be computed, without any hint on how to actually obtain the best design. The core motivation behind my research is to provide autonomous embodied agents with both ef?cient design and robust (self-)adaptation capabilities. In this scope, several aspects of my work are presented, starting with a focus on some important evolutionary mechanisms for automatic design of embodied agents (representation formalism and selection pressure), then addressing the problem of evolutionary design of structural design such as bridges and metallic truss structures, with a special focus on the relevance of developmental design with regards to robustness and scalability. Finally, on-line behavior learning for self-adaptation in the scope of a real-world autonomous robotic agent is described as well as perspectives towards
self-adapting swarm of autonomous agents.

CAUSAL LEARNING FOR DIAGNOSTIC SUPPORT


CAUSAL UNCERTAINTY QUANTIFICATION UNDER PARTIAL KNOWLEDGE AND LOW DATA REGIMES


MICRO VISUALIZATIONS: DESIGN AND ANALYSIS OF VISUALIZATIONS FOR SMALL DISPLAY SPACES
The topic of this habilitation is the study of very small data visualizations, micro visualizations, in display contexts that can only dedicate minimal rendering space for data representations. For several years, together with my collaborators, I have been studying human perception, interaction, and analysis with micro visualizations in multiple contexts. In this document I bring together three of my research streams related to micro visualizations: data glyphs, where my joint research focused on studying the perception of small-multiple micro visualizations, word-scale visualizations, where my joint research focused on small visualizations embedded in text-documents, and small mobile data visualizations for smartwatches or fitness trackers. I consider these types of small visualizations together under the umbrella term ``micro visualizations.'' Micro visualizations are useful in multiple visualization contexts and I have been working towards a better understanding of the complexities involved in designing and using micro visualizations. Here, I define the term micro visualization, summarize my own and other past research and design guidelines and outline several design spaces for different types of micro visualizations based on some of the work I was involved in since my PhD.