PhD Candidate in Machine Learning for Behavior Prediction and Energy Saving in Smart Homes

Climate change is among the most impactful problems of this era. In this project, we aim at developing methods to facilitate a more sustainable life in residential smart homes, supported by the Internet-of-Things (IoT) and Artificial Intelligence (AI). More specifically, we aim at designing machine learning systems to support IoT-enabled smart homes by predicting the inhabitant behavior and energy consumption/saving activities.  

The objectives of the project are (1) the development of a machine learning system that uses the sensor data originating from smart-homes for inhabitant behavior/activity detection and prediction, (2) developing energy-saving models that optimize the inhabitant comfort and energy consumption based on the predicted behaviors of the inhabitants, and (3) developing a digital twin that can visualize the sensor data and simulate scenarios. The PhD Candidate will work alongside domain experts (data science, IoT engineers, and behavioral researchers) and partner companies. There will be access to living-labs in residential households for data collection and prototype-testing. 

Tasks 

• Survey the state-of-the-art machine learning practices in the context of smart homes, sensor networks, IoT, energy consumption optimization, digital twins, and cyber-physical systems;
• Develop algorithms and train/test machine learning models for occupancy/behavior prediction, predictive maintenance, and energy consumption/saving;
• Modeling agent behavior with reinforcement learning, graph neural networks, time series prediction, taking into consideration data and model security and privacy issues and practices;
• Deploy and test prototypes on digital twins and physical living labs; 
• Publish in peer-reviewed relevant scientific journal and conference proceedings;
• Present progress in academic conferences and relevant industry workshops;
• Actively participating in the joint research meetings of the project team and the meetings of the institute of employment;
• Participating for 10- 20% of his/her time in educational activities within the faculty;
• Engaging in valorisation activities.

The candidate will be positioned at the Computer Science department of the Faculty of Science of the Open University. 

Requirements

• MSc in one of the fields (or another related discipline): computer science, machine learning, artificial intelligence, data science, electric/electronics engineering, or a related field with a strong focus on artificial intelligence, data science, data analytics, IoT; 
• A general understanding of security and privacy issues and field practices, e.g., possible threats and GDPR compliance;
• Very good understanding of artificial intelligence, machine learning, and data analytics; 
• Experience in data analysis and developing, training, testing, deploying machine learning systems in Python and/or JavaScript;  
• Good social skills; 
• Ability to work with diverse stakeholders, e.g., industry professionals, academic researchers, living-lab inhabitants; 
• Ambition to collaborate with national and international partners and to stay abroad for (short) research visits in the context of the project; 
• Ambition to develop oneself as a scientific researcher; 
• Fluent written and oral communication skills in English; 
• Motivation to learn Dutch is a bonus;
• Analytical thinking and creativity. 

Salary
The salary is determined in accordance with salary scale P of Appendix A of the Collective Labour Agreement of Dutch Universities and amounts to € 2.770,= gross per month upon commencement, in case of full employment.

The PhD candidate will be appointed for a period of 15 months. The appointment will be extended to 4 years when progress and performance are good.

Station
Heerlen. In consultation with your supervisors, you work on campus or from home.

Flexible studying anywhere in the Netherlands and (Belgium) Flanders
The Open Universiteit is the part-time university in the Netherlands. Students follow personalised and activating academic distance education and disciplinary research is carried out within the various fields of science. Students can complete bachelor and master programmes, but also shorter programmes. The characteristics of education are openness, flexibility and quality (see www.ou.nl/rankings). The Open Universiteit has over 17,000 students and more than 850 employees. The OU has 16 learning centers in the Netherlands and Belgium (see www.ou.nl/studiecentra). The main office is located in Heerlen.

The latest technologies and educational insights are applied both in the bachelor's and master's programmes and courses and in projects and programmes with partners. Nationally and internationally, the OU plays an important role in the innovation of higher education. Education is interwoven with research, which also ensures that the current state of science is incorporated. The Open Universiteit invests not only in disciplinary research in nine scientific fields, but also in research in a multidisciplinary programme: Innovating for resilience.

The faculty of Science is one of the six faculties of the Open Universiteit. Education, research and valorisation are the main tasks. The faculty offers academic bachelor and master programmes in the fields of Computer Science, Information Science, and Environmental Sciences, including a recently developed Artificial Intelligence master programme (started February 2022). The faculty’s research programme focuses on Innovating for Resilience. The faculty has Ph.D.-students who conduct research on various current topics within the research programme. As a result of the interdisciplinary focus on education and research, as well as the close link with practice, work at our faculty provides many innovative and challenging opportunities for entrepreneurial researchers.

Department of Computer Science
The Department of Computer Science at the OU of the Netherlands is an ambitious and enthusiastic group of approximately 40 people (33 fte) focussing on software quality, security and privacy issues, artificial intelligence, and educational and learning technology. The group contributes to these challenges through excellent research and online teaching programmes, covering a large diversity of topics, e.g. automated software testing, cyber security, privacy-aware technologies, deep learning applications in industry, intelligent tutoring systems, and e-health applications. The new chair of Cyber security resides within the Computer Science group and will steer the subgroup of researchers and teachers working on topics related to Cyber security.