Sort by
Refine Your Search
-
Listed
-
Category
-
Country
-
Employer
-
Field
-
to machine learning algorithms in order to get uncertainty estimates for parameters governing the distribution of the observed data. The predictive Bayes scheme for uncertainty quantification contains a wide
-
operationally safe position in real-time. This research focuses on real-time multi-objective optimization of wells, that may be achieved with a mixture of algorithmic and machine-learning approaches. Updating
-
international collaborators across clinical, academic, and industry settings to develop privacy-preserving machine learning approaches, federated learning frameworks, and interpretable algorithms for multimodal
-
National Lab, University of Tokyo etc.), the PhD candidate is expected to research on some of the following themes: New algorithms for parallel/distributed AI/ML Hardware-aware and resource-efficient
-
quantification, in particular the theory and methods known as predictive Bayes. Predictive Bayes theory involves getting Bayesian type uncertainty for parameters given data (i.e., a posterior type distribution
-
prototypes. Signal Acquisition & Processing Develop acquisition pipelines to capture vascular motion during CPR. Implement signal‑processing algorithms to extract markers of blood flow. Model Development
-
collaborators (Simula, Inria, Lawrence Berkeley National Lab, University of Tokyo etc.), the PhD candidate is expected to research on some of the following themes: New algorithms for parallel/distributed AI/ML
-
distributed energy resources (DERs). Design & develop optimization algorithms/tools to plan the deployment of DERs such as energy storage systems (ESS), photovoltaic generations (PV), electric vehicle charging
-
inclusive place to study and work. Postdoctoral Research Fellow position within acoustics and distributed fibre optic sensing At the Department of Physics and Technology , there is a vacancy for a
-
for parameters given data (i.e., a posterior type distribution over the parameter space) without specifying a model nor a prior. Such methods can in principle be applied to machine learning algorithms in order to