Volker Stolz
2018-11-19 12:06:08 UTC
From the advertisement: https://www.jobbnorge.no/ledige-stillinger/stilling/160251/phd-research-fellow-in-reliable-machine-learning
Application deadline: 9. December
Western Norway University of Applied Sciences (HVL), Faculty of Engineering and Science has an open position for a PhD research fellow (PhD position) in Reliable Machine Learning for a period of 4 years.
About the PhD project
Machine learning is changing the landscape of engineering and its application is being still more widespread in society. In particular, software implementations of machine learning algorithms is playing an important role in safety critical systems, including autonomous vehicles, medical diagnostics, and smart software systems.
The correctness of machine learning approaches relies heavily on the training data and the underlying stochastic optimization algorithms. Important elements in order to ensure that systems are based on reliable machine learning therefore includes: ability to control the uncertainty arising when a deployed system is exposed to real data that is far away from the testing data (generalisation problem); ability to obtain reliable prediction despite the stochastic nature of the optimization.
The research focus of the position will be to explore how MODELLING- and SOFTWARE VERIFICATION techniques can be used to improve and assess the reliability of machine learning algorithms. The prospective PhD student will work in close cooperation with our current PhD students working in graph algorithms, software modelling and verification, applied mathematics, and in machine learning.
Application deadline: 9. December
Western Norway University of Applied Sciences (HVL), Faculty of Engineering and Science has an open position for a PhD research fellow (PhD position) in Reliable Machine Learning for a period of 4 years.
About the PhD project
Machine learning is changing the landscape of engineering and its application is being still more widespread in society. In particular, software implementations of machine learning algorithms is playing an important role in safety critical systems, including autonomous vehicles, medical diagnostics, and smart software systems.
The correctness of machine learning approaches relies heavily on the training data and the underlying stochastic optimization algorithms. Important elements in order to ensure that systems are based on reliable machine learning therefore includes: ability to control the uncertainty arising when a deployed system is exposed to real data that is far away from the testing data (generalisation problem); ability to obtain reliable prediction despite the stochastic nature of the optimization.
The research focus of the position will be to explore how MODELLING- and SOFTWARE VERIFICATION techniques can be used to improve and assess the reliability of machine learning algorithms. The prospective PhD student will work in close cooperation with our current PhD students working in graph algorithms, software modelling and verification, applied mathematics, and in machine learning.