Postdoctoral position in Computing Science with focus on Explainable AI and Machine Learning
Job posting number: #7059733 (Ref:AN 2.2.1-1174-19)
Posted: August 25, 2019
Application Deadline: September 16, 2019
Job DescriptionThe Department of Computing Science at Umeå University (http://www.cs.umu.se) is seeking a post-doctoral researcher with focus on Explainable AI and Machine Learning. Autonomous Systems are an important target area. Explaining AI signifies developing new methods for explaining and justifying decisions, recommendations and actions of Artificial Intelligence systems to different users and stakeholders. Explanations should be adapted to the context and to the targeted user. Autonomous Systems signifies autonomously driving vehicles, smart houses, smart machines etc, which interact with the surrounding world and therefore also need to handle different situations (contexts) successfully. It is also often useful or necessary to explain un-successful decisions or failures, especially if there is a risk for injuries to humans.
The recruited post-doctoral researcher will be a member of the Explainable AI team. The team’s high-level vision is to develop ground-breaking Data Science methods for Autonomous Systems that learn from data and experience and that can justify and explain their behavior. This kind of systems are sometimes described using the term Self-* capable Intelligent Products and Environments, where Self-* signifies self-configuring, self-organizing, self-tuning, self-healing, self-managing systems, and self-explaining systems. Future work is expected to have a significant role in the development of such Self-* systems through IoT-enabled remote monitoring and updating over standardized interfaces, together with novel machine learning methods and algorithms are the main technical enablers of such systems.
Your research will focus on one or several of the following key topics:
• Modeling and learning of user and decision-maker actions and preferences in different contexts using non-linear machine learning methods such as neural networks. A challenge with such models is how they can justify their actions and recommendations, and how robust they are in different situations. This is particularly relevant and challenging for autonomous systems whose “intelligence” is based on machine learning and is therefore not directly explainable and understandable by humans.
• Use machine learning and related technologies targeted towards self-learning and adaptive control. Emphasis would be on augmenting the product intelligence embedded in products so that they can auto-adapt themselves according to their operating conditions.
• Autonomous agents and multiagent systems. Intelligent products need to communicate with other products and adapt to them, as well as access and use external information systems that are relevant to them.
The research will be performed jointly with PhD students and senior researchers within our research group, as well as with our national and international partners. Coordination tasks within large-scale projects may also be included.
The appointment is for two years full-time employment. The employment will start November 1,2019, or as otherwise agreed. The position comes with a competitive salary and (non-mandatory) opportunities for teaching (up to 20 % of full-time).
Applicants must have earned a PhD or a foreign degree that is deemed equivalent to a PhD in Data Science or a related subject relevant for the position. The PhD degree should not be more than three years old by the application deadline unless special circumstances exist.Demonstrated experience of artificial intelligence and machine learning is required. Ideal candidates should be capable of independent research and like to work with challenging problems and innovative solutions. Proficiency in English, both spoken and written, is required. Publications in the area of the position are highly meriting. Demonstrated skills in cognitive sciences, psychology, or software development are also considered merits.