Sort by
Refine Your Search
-
architectures and principles from Bayesian neural networks and biological sequence models, including large DNA and protein language models. The project also aims to develop a prototype federated learning
-
Bayesian belief networks; Experience in scenario development approaches, e.g. SSPs; Experience in the application of R-based analytical tools for qualitative or semi-quantitative modelling, incl. RQDA
-
Master Thesis on Bayesian Optimization of Multi-stage Processes with Smart Inducing Point Allocation
, intermediate outputs). This information is highly valuable but not easy to use. With our research we have developed probabilistic methods, which we call Partially Observable Gaussian Process Network (POGPN
-
, intermediate outputs). This information is highly valuable but not easy to use. With our research we have developed probabilistic methods, which we call Partially Observable Gaussian Process Network (POGPN
-
different research directions: Optimization and machine learning with quantum-inspired methods like tensor networks Characterization of the problem graphs resulting from encoding combinatorial optimization
-
or http://github.com/modsim. Quantifying the activity of enzymes operating within the large-scale biochemical network is a fundamental challenge in Systems Bio(tech)nology. Here, the unknown parameters must
-
the possibility of an extension. TASKS: Mathematical modeling and development of inverse methods (e.g. Bayesian inversion, optimization based methods, sparsity promoting methods based on L1-norm minimization and
-
, adversarial attacks, and Bayesian neural networks. Excellent analytical, technical, and problem-solving skills Excellent programming skills in Python and PyTorch including fundamental software engineering