Ph D student, Computing Science, ethical principles of AI systems - Befattningen är under tillsättning eller redan tillsatt, och det är inte längre möjligt att lämna in en ansökan

Research topic: Formal verification of ethical principles of AI systems

Umeå University, the Department of Computing Science, is seeking outstanding candidates for a Wallenberg Autonomous Systems Program (WASP) PhD student position in Computer Science with focus on formal verification of ethical principles of AI systems. Deadline for application is January 31th, 2019.

The Department of Computing Science is a dynamic environment with around 100 employees representing more than 20 countries worldwide. We conduct education and research on a broad range of topics in Computing Science. The Social and Ethical Artificial Intelligence group conducts research into AI systems that meet their social responsibility, including the development of theories, models and tools for oversight, accountability and verification, and methods to analyze, design, measure and audit social implications. The group collaborates with several national and international research groups, and has active links with industry and policy organizations. Visit http://people.cs.umu.se/virginia/ for more information.

WASP
The project is part of the Wallenberg AI, Autonomous Systems and Software Program (WASP), Sweden’s largest ever individual research program, and a major national initiative for strategic basic research, education and faculty recruitment. The program is initiated and generously funded by the Knut and Alice Wallenberg Foundation (KAW) with 2.6 billion SEK. In addition to this, the program receives support from collaborating industry and from participating universities to form a total budget of 3.5 billion SEK. Major goals are more than 50 new professors and more than 300 new PhDs within AI, Autonomous Systems and Software. The vision of WASP is excellent research and competence in artificial intelligence, autonomous systems and software for the benefit of Swedish industry. For more information about the research and other activities conducted within WASP please visit http://wasp-sweden.org/.

The graduate school within WASP provides foundations, perspectives, and state-of-the-art knowledge in the different disciplines taught by leading researchers in the field. Through an ambitious program with research visits, partner universities, and visiting lecturers, the graduate school actively supports forming a strong multi-disciplinary and international professional network between PhD students, researchers and industry. It thus provides added value on top of the existing PhD programs at the partner universities, providing unique opportunities for students who are dedicated to achieving international research excellence with industrial relevance.

Research project
AI systems are increasingly expected to act autonomously. Using current Machine Learning approaches that focus on pattern matching and rely heavily on correlation methods leads to impenetrable systems that are notoriously difficult to monitor (the so called black box algorithms). However, trust on autonomous systems depends on the possibility to verify and minimally control their behavior. In particular, it becomes increasingly important to formally ensure that the behavior of an AI system lays within some given principles. 

An important research area in AI is therefore the development of theories and methods for control and verification of AI systems that adapt to their environment without direct intervention of the users. This requires going beyond traditional verification and monitoring methods, for two main reasons. Firstly, these systems operate in open environments, which cannot be fully described nor defined a priori, and therefore rules and constraints on its behavior cannot consider all possible situations encountered by the system. Secondly, opacity is inherent in open environments because the systems are developed and operated by different parties, using that use different (proprietary) architectures and algorithms.

However, trust in the system requires that we are able to ensure that the system complies with societal values and ethical principles, such as fairness, non-discrimination, safety or privacy. Theories and tools are needed to specify computational representations of these concepts, and to verify that a given system complies with these. In this project, you will develop methods to verify and monitor the ethical behavior of AI systems based on the observation of input and output behavior according to a continuously evolving societal optimum.

About the position
The successful applicant will be employed by Umeå University and receive a competitive salary for a period of four years of full time PhD studies, provided that the studies are progressing reasonably well. PhD students are typically offered the opportunity to gain teaching experience on suitable undergraduate courses. If so, the employment period is extended to cover the time spent on teaching (at most 20 %). Expected starting date is March 1 2019 or as otherwise agreed.

Admission requirements
The applicant is required to have completed a second- cycle level degree, or completed course requirements of at least 240 ECTS credits, of which at least 60 ECTS credits are at second-cycle level, or have an equivalent education from overseas, or equivalent qualifications. To fulfil the specific entry requirements for doctoral studies in computing science, the applicant is required to have completed courses at second-cycle level degree equivalent to 60 ECTS credits in computing science, or in a subject considered to be directly relevant for the specialization in question.

Candidates are expected to have very good knowledge in Artificial Intelligence and formal methods. Demonstrable knowledge of logics, theorem proving and verification is a prerequisite. In particular, candidates should be well acquainted with relevant formal and algorithmic techniques, and with knowledge representation methods which, as the project proceeds, will be applied to develop verification and monitoring models for AI algorithms.

Since research is conducted in an international research environment, the ability to collaborate and contribute to teamwork, and a very good command of the English language, both written and spoken, are key requirements.

Application
A complete application should contain the following documents:
• A cover letter including a description of your research interests, your reasons to apply for the position, and your contact information
• A curriculum vitae
• If applicable, copy of completed BSc and/or MSc thesis and other original research publications
• Copies of degree certificates, including documentation of completed academic courses and obtained grades
• Contact information for two persons willing to act as references

The application must be written in English or Swedish, and attached documents must be written in or translated to English or Swedish. Documents must be in Word or pdf format. Applications must be submitted electronically using the e-recruitment system of Umeå University, and be received no later than January 31 th, 2019. Reference number: AN 2.2.1-2063-18. Log on to the system and apply via the button at the bottom of the web page.

Interested applicants are also encouraged to apply for the position with the reference number: AN 2.2.1-2063-18.

The procedure for recruitment for the position is in accordance with the Higher Education Ordinance (chapter 12, 2§) and the decision regarding the position cannot be appealed. We value the qualities that an even gender distribution brings to the department, and therefore we particularly encourage women to apply for the position.
For additional information, please contact Prof. Virginia Dignum virginia@cs.umu.se.

We look forward to receiving your application!