A5053456353- Model Reduction and Neural Networks
- Urban and Freight Transport Logistics
- Neural Networks and Applications
- Vehicle emissions and performance
- Stochastic processes and financial applications
- Advanced Numerical Methods in Computational Mathematics
- Fluid Dynamics and Turbulent Flows
- Computational Physics and Python Applications
- Advanced Mathematical Modeling in Engineering
- Consumer Retail Behavior Studies
- Gaussian Processes and Bayesian Inference
- Transportation Planning and Optimization
- Meteorological Phenomena and Simulations
- Transport Systems and Technology
- Climate Change Policy and Economics
- Anomaly Detection Techniques and Applications
- Time Series Analysis and Forecasting
- Spacecraft Dynamics and Control
- Urban Transport and Accessibility
- Traffic Prediction and Management Techniques
- Merger and Competition Analysis
Chemnitz University of Technology
2019-2021
Bicycle usage is significantly affected by weather conditions. Climate change is, therefore, expected to have an impact on the volume of bicycle traffic, which important factor in planning and design infrastructures. To predict traffic a changed climate city Berlin, this paper compares traditional statistical approach three machine learning models. For purpose, cross-validation procedure developed that evaluates model performance basis prediction accuracy. XGBoost showed best used for...
Second-order partial differential equations in non-divergence form are considered. Equations of this kind typically arise as subproblems for the solution Hamilton-Jacobi-Bellman context stochastic optimal control, or linearization fully nonlinear second-order PDEs. The these problems is natural. If coefficients diffusion matrix not differentiable, problem can be transformed into more convenient variational form. We investigate tailored non-conforming finite element approximations PDEs form,...
Abstract Second‐order partial differential equations (PDEs) in nondivergence form are considered. Equations of this kind typically arise as subproblems for the solution Hamilton–Jacobi–Bellman context stochastic optimal control, or linearization fully nonlinear second‐order PDEs. The these problems is natural. If coefficients diffusion matrix not differentiable, problem cannot be transformed into more convenient variational form. We investigate tailored nonconforming finite element...
We present a novel technique based on deep learning and set theory which yields exceptional classification prediction results. Having access to sufficiently large amount of labelled training data, our methodology is capable predicting the labels test data almost always even if entirely unrelated data. In other words, we prove in specific setting that as long one has enough points, quality irrelevant.
In this paper we focus on comparing machine learning approaches for quantum graphs, which are metric i.e., graphs with dedicated edge lengths, and an associated differential operator. our case the equation is a drift-diffusion model. Computational methods require careful discretization of operator that also incorporates node conditions, in Kirchhoff-Neumann conditions. Traditional numerical schemes rather mature but have to be tailored manually when becomes constraint optimization problem....