Postdoctoral Computer Science, Context-Based Machine Learning in Autonomous Systems Umeå University, Department of Computer Science

Postdoctoral computer science, context-based machine learning in autonomous systems


Department of Computer Science

 

The Department of Computer Science at Umeå University (http://cs.umu.se) is now seeking an excellent candidate for postdoctoral position in contextual machine learning for autonomous systems, including methods of artificial intelligence methods for explaining and motivating the learned behavior.


Deadline for application is 2018-01-08.


The position is associated with Wallenberg Autonomous Systems and Software Program (WASP, http://wasp-sweden.org/). WASP is Sweden's largest individual research program ever and offers a platform for academic research and education that interacts with Swedish leading industry. Within the program, research is conducted on autonomous systems that interact with people and adapt to the current environment using sensors, information and knowledge in an intelligent system of systems. WASP's core values are the research sex and industrial relevance.


Project description 


research group has a vision to develop pioneering methods for autonomous systems to learn from data and experience, as well as to explain and motivate their decisions. A term that can be used to describe such systems is Self-*, which means systems capable of configuring themselves, organizing themselves with other systems, adjusting their own behavior, diagnosing and repairing themselves, and administering themselves for payment. , software updates, call for external help, etc. The project is expected to advance the development of such Self-* systems, especially in the so-called Internet of Things (IoT) context where information from donors, control systems, users and other systems forms an intelligent system of systems where different systems can communicate over standardized protocols and thus also maintain self-capability as they interact in larger systems.


Your research will be focused on one or more of the following areas:


Develop and apply reward-based learning (reinforcement learning) and other learning methods for self-learning and self-optimizing control. Increasing memory and counting capacity can be built into different systems, which can thus learn and adapt to different situations and conditions. In addition, embedded systems can communicate with other systems (built-in, in "cloud" or elsewhere) and connect with them for more comprehensive information and to optimize the functionality of larger systems.


Autonomous agents and systems of agents to create so-called "cyber-physical systems" (CPS) where different systems, including complex ones, can control and control their function under the so-called System of Systems (Systems of Systems, SoS; see t .ex. https://ec.europa.eu/digital-single-market/en/system-systems) principles. CPS also takes into account the human factor and society in general.


Analysis and learning of models for how people make their decisions in different situations. For example. Neural networks can be used for such analysis. However, the challenge with neural networks and other machine learning technologies is to explain and motivate why a decision has been made or an action has been taken in a specific situation. Such methods are particularly important in explaining and motivating decisions and actions made by autonomous systems whose "intelligence" is entirely based on machine learning and therefore can not be explained or understood by human users.


The research will be carried out in collaboration with senior researchers and doctoral students in their own research group and with our national and international external partners. Project coordination within large-scale projects may also be included in the service.


The employment covers full-time (100%) for 2 years. Start 2018-04-01 or by appointment. The service has a market salary and there is the opportunity to teach up to 20%.


Qualification


requirements Qualifying to be employed as a postdoctoral degree is the one who has obtained a doctorate or a foreign degree which is deemed to be equivalent to a doctorate in computer science or other relevant subject area. The exam must be completed no more than three years before the expiry of the application deadline unless special reasons exist.


Good insight into artificial intelligence, machine learning and distributed multi-agent systems is a requirement. Research essays and excellent scientific publications in the field of employment are highly credible. Documented knowledge and experience of cognitive sciences, psychology and software development is meritorious. The applicant should also be driven by a strong research interest, well-organized, and enjoying working with challenging problems and innovative solutions. Very good knowledge of spoken and written English is a requirement.


Application


A full application must contain:

 

  • A cover letter with a map (about 2 pages) Description of your research interests in relation to the above-mentioned research area, and how you could contribute to the project
  • Curriculum Vitae (CV), with a complete publication list
  • Copies of diplomas and so on
  • A copy of your doctoral dissertation, as well as copies of relevant scientific publications (max 5), numbered according to the publication list
  • Contact information for three reference persons


Other relevant information, such as documentation of software development experience or work in or with industry.


All material must be in Swedish or English. If materials are submitted in other languages, a translation must be included in Swedish or English.


Welcome with your application! The application must be made via the MyNetwork Pro e-recruitment system, and be received by 2018-01-08. Ref No: AN 2.2.1-1807-17.


More information


For questions or needs for further information, contact (primarily) Professor Kary Främling, Kary.Framling@cs.umu.se, phone: + 358- (0) 50-5980451.


More information about Umeå University is available at http://www.umu.se


Type of employment
probationary


Extent
100%


Salary
Monthly Salary


Contact
Kary Främling, Professor
Tel: +358 (0) 50-5980451


Union contacts
SACO
Tel: 090-786 53 65


SEKO
Tel: 090-786 52 96


ST
Tel: 090-786 54 31


Access
2018-04-01 or by agreement


Reference number
AN 2.2.1-1807-17


If you apply for this position please say you saw it on Computeroxy

Apply

All Jobs

surrey

ubc reklama

Anu

cambridge

geneva

kuwait

Melbourne

nottingham

Nus

sfu

southampton

texas tech

Toronto

uni copenhagen

unsw

Uwo