Two PhD and two postdoc positions in logic for AI - Utrecht University

[Jan Broersen]

Two PhD and two postdoc positions in logic for AI - Utrecht University

Deadline for application: February 1st, 2014

For the ERC consolidator project "Responsible Intelligent Systems" 
(REINS) we are looking for 2 PhD students and 2 postdocs to be employed 
in the division of artificial intelligence at the department of 
information and computing sciences.

REINS

The REINS project aims to provide a framework for automating 
responsibility, liability, and risk checking for intelligent systems. As 
intelligent systems are increasingly integrated into our daily life, the 
division, assignment and checking of responsibilities between human and 
artificial agents become increasingly important. From robots in 
medicine, the military and transportation (self-driving cars); to 
automated trading agents on the financial markets (algotrades); to 
automated monitoring, prediction and protection systems (pacemakers, 
automated surveillance, early warning systems); we delegate more and 
more responsibility to intelligent devices. By delegating 
responsibilities to intelligent devices, we run the risk of losing track 
of our indirect legal and moral liabilities.

Sub-project 1: "Logics of Collective Responsibility"

Vacancies: 1 PhD student (4 years), 1 postdoc (2 or 3 years)
Starting date: March 1st, 2014 (later dates may be negotiated)
Goal: Suitable logical representation languages for individual and 
collective responsibilities.
Scientific context: computer science, philosophy, logic, artificial 
intelligence

The "Logics of Collective Responsibility" sub-project undertakes the 
formalization of the concept of responsibility, with a focus on the 
generalization of existing formal theories to the collective case. The 
concept of responsibility is highly non-trivial. It has been the subject 
of extensive debates in philosophy and legal theory and revolves around 
issues like free will, moral luck and collective action. In our opinion 
the conceptual and philosophical complexities involved in the 
understanding of the notion of responsibility ask for the clarity and 
expressivity of logical formalizations. While for the case of individual 
responsibility several formal languages have been proposed, for the case 
of collective responsibility, the logic literature is lacking concrete 
suggestions. The key objective in this sub-project is to fill this 
knowledge gap.

Sub-project 2: "Grades of Responsibility relative to Normative Systems"

Vacancies: 1 PhD student (4 years), 1 postdoc (3 years)
Starting date: March 1st, 2014 (later dates may be negotiated)
Goal: A logic theory of grades of responsibility relative to normative 
systems
Scientific context: computer science, legal theorie, deontic logic, 
artificial intelligence

In the "Grades of Responsibility relative to Normative Systems" 
sub-project we will develop systems of formal deontic logic in relation 
to classes of responsibility (as developed in sub-project 1) and 
probabilistic, possibilistic or evidence-based action. The possibility 
to model actions and events using probabilities has been investigated in 
the area of Markov Decision Processes and in Bayesian Networks. However, 
in the context of deontic reasoning, actions with uncertain effects have 
not been considered. Yet it is very natural to think of having 
responsibilities relative to a normative system as having to optimize 
the chance to obey obligations and having to avoid the risk of 
violations. We can even make a stronger claim: responsibility typically 
only shows in normative contexts where uncertainty and risk play a 
central role. The project aims to combine existing theories on 
responsibility, risk-taking action and deontic logic with existing 
theories of probabilistic, possibilistic or evidence-based reasoning.

*Utrecht University*

Utrecht University has great ambitions for its teaching quality and 
study success rates. This also applies to its clear research profiles 
which are centred around four themes: Sustainability, Life Sciences, 
Youth & Identity, and Institutions. Utrecht University plays a prominent 
role in our society and contributes to finding the answers to topical 
and future societal issues.

*Faculty of Science*

The Faculty of Science consists of six departments: Biology, 
Pharmaceutical Sciences, Information and Computing Sciences, Physics and 
Astronomy, Chemistry and Mathematics. The Faculty is home to 3500 
students and nearly 2000 staff and is internationally renowned for the 
quality of its research. The Faculty's academic programmes reflect 
developments in today's society.

*Department of Information and Computing Sciences*

The Department of Information and Computing Sciences is nationally and 
internationally well known for its fundamental research in computer 
science and as an initiator in the area of game technology. The 
department's research activities are clustered into four divisions, viz. 
Virtual Worlds, Artificial Intelligence, Interaction Technology, and 
Software Systems. The Department of Information and Computing Sciences 
offers educational bachelor programs in computer science and information 
science, and four (English) research master programs, viz. Game and 
Media Technology, Computing Science, Business Informatics, and 
Artificial Intelligence. Three years ago the Department introduced a 
track on Game Technology in the bachelor program in computer science, 
which led to a substantial annual increase of bachelor students. The 
bachelor programs are the largest computer science and information 
science programs in the Netherlands

*Division Artificial Intelligence*

The division Artificial Intelligence comprises the chairs Intelligent 
Systems, Decision Support Systems and Algorithmic Data Analysis. 
Together these groups cover a broad spectrum of subjects in Artificial 
Intelligence, ranging from symbolic logic approaches to sub-symbolic 
evolutionary approaches; from agent programming to machine learning; 
from knowledge discovery to knowledge representation; from Bayesian 
networks to applied algorithmics for AI.

*Job Description*

Candidates are expected to:
- complete and defend a PhD thesis within the nominal time-frame 
(applies only to PhD candidates);
- assist in the supervision of the PhD student in the sub-project 
(applies only to postdocs);
- regularly present intermediate research results at our internal 
colloquium and at international workshops and conferences;
- collaborate with the other researchers in the REINS project;
- participate in the organization of research activities and events such 
as conferences, workshops and symposia;
- Assist in teaching tasks and/or take on teaching responsibilities

*Conditions of employment*

PhD candidates are offered a full-time position for 4 years. Salary 
starts at € 2,042.- and increases to € 2,612.- gross per month in the 
fourth year of the appointment. The salary is supplemented with a 
holiday bonus of 8% and an end-of-year bonus of 8,3% per year. In 
addition we offer: a pension scheme, a partially paid parental leave, 
flexible employment conditions. Conditions are based on the Collective 
Labour Agreement Dutch Universities. More information is available on: 
http://www.uu.nl/EN/informationfor/jobseekers/Working-for-Utrecht-University/terms-of-employment/Pages/default.aspx

Postdoc candidates are offered a position for 2 or 3 years. The 
positions are full-time, but part-time (80%) employment is negotiable. 
The gross salary depends on qualifications and experience and ranges 
from € 2,919 to € 3,831 per month. (Salary scale 10) The salary is 
supplemented with a holiday bonus of 8% and an end-of-year bonus of 8,3% 
per year. In addition we offer: a pension scheme, a partially paid 
parental leave, flexible employment conditions. Conditions are based on 
the Collective Labour Agreement Dutch Universities. More information is 
available on: 
http://www.uu.nl/EN/informationfor/jobseekers/Working-for-Utrecht-University/terms-of-employment/Pages/default.aspx

*How to apply*

Candidates are invited to send (1) a motivation letter, (2) a CV, (3) 
the email address of a referee, and (4) a published paper (for postdocs) 
or a master thesis plus grades (for PhD candidates) by following the 
instructions on the vacancies page of our faculty 
http://www.uu.nl/faculty/science/EN/AboutFBW/organisation/Pages/vacancies.aspx 
or on the vacancies page of our departement 
http://www.cs.uu.nl/vacatures/en/ Deadline for applications is february 
1st, 2014. For any further information on the project or the vacancies 
one can contact dr. Broersen at email address J.M.Broersen at uu.nl or 
professor John-Jules Charles Meyer at email address J.J.C.Meyer at uu.nl