Sort by
Refine Your Search
-
Category
-
Field
-
minimizing error and maximizing efficiency, is computationally challenging—no known polynomial-time algorithm exists to solve it optimally in all cases. Because of this complexity, researchers typically rely
-
p x∈R y∈Rp where F is the outer objective and f is the inner objective. Solving such problems is challenging due to the need to compute gradients through
-
suitable data models [CSC+23]. Objectives As far as the design of efficient numerical algorithms in an off-the-grid setting is concerned, the problem is challenging, since the optimization is defined in
-
on stochastic Riemannian optimization algorithms, these methods still suffer from limitations in computational complexity. The post-doctoral fellow will build upon this preliminary work to investigate
-
algorithms for dynamic structured data, with a particular focus on time sequences of graphs, graph signals, and time sequences on groups and manifolds. Special emphasis will be placed on non-parametric
-
at the interface of machine learning and computational neuroscience. The candidate will be part of the COATI joint team between INRIA d’Université Côte d’Azur and the I3S Laboratory. Project The candidate should
-
revisit discretization methodologies in view of modern requirements and computational capabilities. The candidate will focus on developing mesh generation algorithms meeting the following criteria
-
) models. Little attention has been paid to other models like graph NNs (GNNs) or PCA. Among the few existing works in the literature, [2] proposes an FL algorithm to compute PCA in a DP fashion, but the
-
on Neural Information Processing System [20] Anish Agarwal, Munther Dahleh, and Tuhin Sarkar, A marketplace for data: An algorithmic solution, in Proceedings of the 2019 ACM Conference on Economics and
-
, sensor failures, or the aggregation of datasets from multiple sources. There is a rich literature on how to impute missing values, for example, considering the EM algorithm [Dempster et al., 1977], low