Agenda
September
-
14:0017:00
During this defense, I’ll present part of my work on computer graphics applied to the creation of animated films, and more specifically on how digital tools can help artists reproduce the distinctive characteristics of traditional 2D animation, both in terms of appearance and movement. This work is part of a research trajectory that has gradually shifted from expressive rendering to the stylization of form and movement, and finally to assisting in the creation of sketched 2D animations. A common thread runs through this work: placing artistic control at the heart of the design of the proposed solutions, in order to offer artists interactive tools compatible with their professional practices.
Finally, I’ll outline several perspectives on the coherent integration of 2D and 3D content. To achieve this, beyond technical challenges, a central scientific question remains: how to characterize the perceptual cues that distinguish movements in traditional 2D animation from those generated by 3D animations.
FrançaisAmphi LaBRI -
14:0018:00
The Shapley value is a method of wealth distribution defined in the 1950s as the only method that satisfies a set of desirable axioms. In recent years, it has been applied to databases to define measures of responsibility that quantify the contribution of each fact to a given response. These measures—and other similar applications of the Shapley value to different fields—generally have two drawbacks, one conceptual and the other more practical: (1) Shapley’s axioms are invoked to justify them, but these properties—however desirable they may be in economics—are not meaningful in all contexts, and as a result, the resulting measures sometimes exhibit unexpected behavior; (2) Measures based on the Shapley value are often difficult to compute, since in many cases there are very simple queries for which the problem is nonetheless #P-hard, even in terms of data complexity. This thesis extends these accountability measures to queries involving ontologies and addresses these two shortcomings of existing Shapley-value-based accountability measures, both in the context of databases and ontologies.
For the first problem, we reexamine the question of what constitutes a good measure of responsibility for responses to queries; to this end, we identify properties inspired by Shapley’s axioms that are truly desirable in the specific context we are studying.
For the second, we define new measures, which are still based on the Shapley value but are much easier to compute. To do this, we use other “resource functions” to model the instance under study as a cooperative game to which the Shapley value applies. In addition to these conceptual considerations, we study in detail the complexity of all the measures under consideration, focusing primarily on various forms of conjunctive queries—sometimes enriched with unions and negative atoms—as well as ontologies expressed in lightweight description logics from the DL-Lite and EL families. Finally, we draw on the insights into the Shapley value derived from our study of such accountability measures to take a fresh look at other applications of the Shapley value, in particular the SHAP score, a measure widely used in artificial intelligence to explain the results of classifiers.
Amphi LaBRI -
14:0018:00
Every day, BNP Paribas Personal Finance (BNPP PF) receives thousands of requests for installment payments, which are processed in a matter of seconds. This speed is what defines the quality of the service, but it also makes it vulnerable: fraudsters exploit it to obtain financing with no intention of repaying it. To detect these cases, BNPP PF links the requests in a graph, where each request becomes a node connected to others that share identifying information. This relational structure is a treasure trove of information that has yet to be fully tapped. This thesis demonstrates how to leverage it to better detect fraud, while meeting the service’s real-time constraints. It makes three contributions. First, a multi-level identity resolution method that reveals links between applications, including a method adapted for split payments. Second, a hybrid architecture combining graph neural networks and ensemble methods to exploit both individual and relational signals. Finally, the publication of an anonymized and distorted dataset derived from real-world data. The results confirm the value of the graph approach and the hybrid coupling, the benefits of which increase when relational information is lacking in the data.
Amphi LaBRI -
14:0018:00
In ontology-mediated query answering, an ontology (often specified using description logics, DLs) is used to enrich incomplete data with domain knowledge, which is taken into account when computing query answers. A central issue in this setting is how to proceed when the data is inconsistent with the ontology, since in classical logic an inconsistent theory entails every formula, so that inconsistency-tolerant semantics are needed to obtain meaningful answers. We consider three existing inconsistency-tolerant semantics based upon the notion of a repair, defined as an inclusion-maximal subset of the data consistent with the ontology. These three semantics can be used conjointly to identify answers with different levels of confidence: those that hold in every repair (AR semantics), those entailed by the facts common to all repairs (IAR semantics), and those holding in at least one repair (brave semantics). The thesis focuses on repair-based and cost-based semantics which take into account preference information reflecting the differing reliability of the facts and/or axioms, given either by a qualitative priority relation between conflicting facts or numerical weights.
Amphi LaBRI