Dr. Hannes Stuke

Department of Mathematics and Computer Science, Institute of Mathematics

Group Nonlinear Dynamics

Group Mathematics of Machine Learning

Principal Investigator

Address
Arnimallee 7, rear building
Room 133
14195 Berlin
Email
h.stuke[at]fu-berlin.de

Research interests

  • Machine learning
  • Dynamical Systems
  • Blow-Up of Evolutionary PDEs
  • Bianchi Cosmological Models
  • Dynamical systems
  • Applications to biology
  • Applications to medicine

Current Projects

Recent Projects

Gurevich P., Stuke H.
Robustness Against Outliers For Deep Neural Networks By Gradient Conjugate Priors
Preprint:  arXiv:1905.08464 [stat.ML]
(pdf)

Stuke H.
Complex time blow-up of the nonlinear heat equation
Preprint arXiv:1812.10707 [math.DS]
(pdf)

Gurevich P., Stuke H. 
Gradient conjugate priors and deep neural networks.
J. Artificial Intelligence 10.1016/j.artint.2019.103184   
(pdf)

Gurevich P., Stuke H.
Pairing an arbitrary regressor with an artificial neural network estimating aleatoric uncertainty.

Neurocomputing. 10.1016/j.neucom.2019.03.031
(pdf)

Gurevich P., Stuke H., Kastrup A., Stuke H. and Hildebrandt H. 
Neuropsychological testing and machine learning distinguish Alzheimer's disease from other causes for cognitive impairment.
 
Front. Aging Neurosci. 9:114 (2017) 
(pdf)
Fila M., Stuke H.
Very slow stabilization for a nonlinear Fokker–Planck equation 
Nonlinear Analysis, Theory, Methods & Applications (2016)
(pdf
Fila M., Stuke H. 
Special asymptotics for a critical fast diffusion equation
 
Discrete and Continuous Dynamical Systems (2014)
(pdf)
Ghoussoub's self-dual variational principle
Bachelor Thesis, Free University Berlin, 2011 
(pdf, 330 kB)