Project: Hybrid Intelligence
Keywords: non-monotonic reasoning, computational logic, user modelling
Offer: fully funded 4-year PhD position at University of Twente, The Netherlands
Deadline for application: February 11th, 2022
Digital information technology is increasingly interwoven with our society and individuals' daily lives. With these digital technologies tracking and influencing our everyday behavior, they should align with the user’s needs and values. To achieve this, personal agents need to model the user’s informational and motivational attitudes such as goals, values, norms, beliefs, and their interrelations in order to derive support actions that are aligned with the user. This is challenging because 1) conflicts may occur in the user’s motivations, and 2) the user’s state and motivations may change over time and in different contexts.
We offer a fully funded 4-year PhD position to investigate how non-monotonic reasoning techniques such as belief revision, argumentation, defeasible reasoning, inconsistency handling, and reasoning about actions can be employed for user modelling in the context of personal agents. Questions that can be addressed are for example: Which types of update operations on user models are required? Is a new norm in conflict with existing desired behaviors? Can belief revision techniques be adapted for run-time norm revision? With this we lay the foundations for flexible and value-aligned personal agents that give people agency over their lives with digital technology.
The position is associated with the Hybrid Intelligence Centre (https://www.hybrid-intelligence-centre.nl), a 10-year Dutch national research program which investigates how we can create AI that augments, instead of replaces human intelligence.
The prospective candidate is expected to perform and publish high quality and internationally visible research, and be involved in teaching and student supervision to a limited extent.
- You hold or will soon hold a Master's degree (or equivalent) in computer science, artificial intelligence, or a related field;
- You have excellent proficiency in English in speaking and writing (C1; above IELTS 7 or equivalent);
- You have good communication skills, take initiative, and have an interest in working in a multidisciplinary environment;
- You have interest and some experience in one or more of the following, or related topics: logic-based agents, knowledge representation and reasoning, formal methods, or non-monotonic reasoning.
WHAT WE OFFER
The position is embedded in the Human Media Interaction (HMI) department at University of Twente. HMI is a friendly and welcoming research group, with attention to employee wellbeing. The terms of employment are in accordance with the Dutch Collective Labour Agreement for Universities (CAO) and include:
- A fulltime PhD position for four years, with a qualifier in the first year.
- Full status as an employee at the UT, including pension and health care benefits.
- The salary will range from € 2.434,- (1st year) to € 3.111,- (4th year) per month, plus a holiday allowance of 8% and a year-end bonus of 8.3%;
- A solid pension scheme;
- A minimum of 29 leave days in case of fulltime employment, based on 38 hours a week. A week schedule of 40 hours would result in 12 extra leave days;
- Excellent facilities for professional and personal development.
HOW TO APPLY
Applications will be accepted until the position has been filled. The first round of interviews will be conducted based on applications received up to and including February 11th, 2022. The preferred starting date is between May 1st 2022 and September 1st 2022. Informal inquiries about this position can be sent to Birna van Riemsdijk (email@example.com).
More information about how to apply, and about the HMI research group and the University of Twente can be found via this link.
Digital information technology is increasingly interwoven with our society and individuals' daily lives. On one hand, a tremendous amount of data becomes available in machine-interoperable formats while how humans can intuitively access and learn from such large scale knowledge is becoming an urgent question. On the other hand, with these digital technologies tracking and influencing our everyday behaviour, they need to align with the user’s needs and values to become truly extensions of our “selves”.
Addressing these challenges requires both research on how to shape knowledge-centered human-machine conversations and meaning-making in an engaging and intuitive manner, as well as fundamental research regarding the formal specification and run-time revision of knowledge structures for modelling users’ norms and values in a personal agent.
We have fully funded positions available on the topics described in more detail below.
Conversational access of large scale knowledge graphs:
The aim is to investigate conversational AI technologies to achieve a meaningful, engaging and enjoyable interaction between individual persons and large scale knowledge graphs. This includes selecting and packaging relevant information into narratives, choosing suitable strategies to convey knowledge. Such interaction needs to take into account the individuals' cognitive and information processing capabilities as well as their personal norms and values.
Knowledge representation & reasoning for value-aligned personal agents:
The aim is to develop formal knowledge representation and reasoning techniques that allow a personal agent to model the user’s (inter)personal norms, habits and underlying values, and revise these “personal knowledge structures” at run-time based on user input. Questions that can be addressed are for example: How to model temporal aspects of daily routines? Which types of update operations on personal knowledge structures are required? Is a new norm in conflict with existing desired behaviours? Can belief revision techniques be adapted for run-time norm revision? With this we lay the foundations for flexible and value-aligned personal agents that give people agency over their lives with digital technology.
For more information about the Postdoc positions and how to apply, please click here.