[Учеба]  PhD Position on Explainable Data Science using Formal Verification @Radboud University, Nils Jansen
Сообщение было послано: qwera (5x18x238x142.static-business.iz.ertelecom.ru)
Дата: Воскресенье, Апрель 14 13:24:40 2019


(repost from the mailing list)

Are you interested in improving the explainability of machine learning using formal verification? Are you interested in being a part of cutting edge research in cooperation with TNO and the iCIS institute at Radboud University, Nijmegen? Then apply for a PhD position within the EXoDuS project, where you will be jointly supervised by Dr. Nils Jansen (Radboud University, Nijmegen) and Dr. Guillermo Perez (University of Antwerp).

* Responsibilities

As a PhD candidate, you will work on the project EXplainable Data Science using Formal Verification (EXoDuS). The project is partially funded by the VWDATA programme and the Institute for Computing and Information Science (iCIS) of Radboud University and will be carried out in close cooperation with TNO.

The impact of artificial intelligence (AI) on autonomous driving, robot-assisted surgery, and home automation has resulted in an increased reliance on AI systems. In mission-critical applications, the inherent vulnerability of such systems to adversarial attacks poses a serious challenge. We propose to immerse humans in the process of robustifying AI systems against problems such as adversarial learning or data poisoning via formal methods. Formal verification in particular provides tools and algorithms to rigorously assess the correctness of a system. The key element is to enable humans to understand AI-made decisions in an adversarial environment. Decision-making is sufficiently captured by so-called strategies; a neural network, for example, represents a strategy that has been learned. For such strategies, data scientists and system engineers lack tools to answer transparency-related questions.

You will use formal techniques and directions such as SMT solving, model checking, and partially observable Markov decision processes. From the AI perspective, you will marry the aforementioned concepts with (convolutional) deep neural networks and decision trees.

* Work environment

Strategically located in Europe, Radboud University is one of the leading academic communities in the Netherlands. The position is available in the Software Science group of the Institute for Computing and Information Sciences (iCIS) at Radboud University. Research at iCIS focuses on software science, digital security and data science. Our research mission is to improve the security and reliability of computer-based systems and algorithms through mathematically founded theories, methods and tools. During recent evaluations, iCIS has been consistently ranked as the No. 1 Computing Science department in the Netherlands. Evaluation committees praised our flat and open organisational structure, our ability to attract external funding, our strong ties to other disciplines, and our solid contacts with government and industrial partners. The Software Science group is well known for its contributions to the mathematical foundations of software, formal methods, and functional programming.

* Details

The application deadline is April 14, 2019. More information and application form: https://www.ru.nl/werken/details/details_vacature_0/?recid=601933

For questions please contact Nils Jansen (n.jansen@science.ru.nl).

* Further information

http://nilsjansen.org

https://www.uantwerpen.be/en/staff/guillermoalberto-perez/

https://www.ru.nl/english/working-at/why-work-at-radboud-university-0/terms-employment/


Сообщения в этом потоке
+ [Учеба]  PhD Position on Explainable Data Science using Formal Verification @Radboud University, Nils Jansen (url) (161) - qwera (5x18x238x142.static-business.iz.ertelecom.ru) - 14/4/2019 13:24
Ответить

Имя:   Пароль:    Автологин
Тема:
        

Отключить распознавание Тегов конференции
Отключить распознавание смайл-кодов
Получать уведомления об ответах по почте

 

Конференция основана на движке WWWConf 2.0 PRE BETA5, поддерживается и модерируется группой энтузиастов.