PhD Research Fellowship Position in Reinforcement Learning for Intelligent, Autonomous IoT Devices
- Full Time
Norwegian University of Science and Technology (NTNU) has a vacancy for a 100% position as a PhD research fellowship position affiliated with the Department of Information Security and Communication Technology (IIK) at Faculty of Information Technology and Electrical Engineering (IE)
Information about the department
The Department of Information Security and Communication Technology has 33 permanent scientific staff members, and in total approximately 80 employees. 50 persons work at NTNUs campus in Gjøvik and 30 at NTNUs campus in Trondheim. IIK carries out internationally competitive research organized in seven cross-disciplinary labs covering the academic areas of Internet of Things (IoT), Applied Cryptology, Intelligent Transport Systems, Quantifiable Performance and Dependability of Communication Systems, Biometry, Digital Forensics, and Information Security. The department provides teaching in BSc in IT Operations and Information Security, MSc programmes in Communication technology and Information Security, and PhD programmes in Telematics and Information Security. The department hosts and operates NTNUs Center for Cyber and Information Security, a public-private partnership with 25 members.
This is a researcher training position aimed at providing promising researcher recruits the opportunity of academic development in the form of a doctoral degree.
Part of the digitization in domains like smart cities, precision agriculture, manufacturing, and healthcare is the instrumentation of the physical world with IoT nodes to gather data, as well as make local changes to improve the domain’s overall performance. In current IoT solutions, decision-making is concentrated on the server side. The logic of IoT nodes is primitive and often static, with parameters set by developers at design-time. Since IoT devices are deployed in environments with changing conditions, such statically chosen parameters are far from optimal. Individual, manual control, however, is not possible due to the scale of the systems and the complexity of factors to take into account. As a result, IoT nodes often fail during operation, deliver much less data than desirable, or need to be equipped with far larger energy sources than ideally necessary. Instead, future IoT devices must act intelligently and independently to serve their application goals optimally given their resource situation. Furthermore, we expect that good solutions are so complex that they cannot be provided in advance of deployment. It is therefore also important that the system enables IoT nodes to learn their behavior autonomously. To accomplish this, the ROBIOT project will apply and evolve deep reinforcement learning to be more appropriate for this scenario.
Recently, deep reinforcement learning has been applied with great success to problems in which agents needed to autonomously learn strategies. Our understanding of IoT nodes indicates that they fit well with the concept of a reinforcement learning agent, since IoT nodes need to make complex decisions to balance between different concerns such as the energy budget, the communication channel, or the application requirements. Learned planning has the capability to increase the efficiency and effectiveness of these ubiquitous devices. Like robots, IoT nodes need to navigate through a series of choices, plan ahead, learn about their environment, and learn which actions have which consequences. For instance, they need to learn how much energy is consumed by each action, how much energy is gained from energy harvesting throughout a day, which communication channels are most effective, and how much data they should deliver to the decision-support system in a given situation. The major research challenge is that deep reinforcement learning depends on a large amount of training data and substantial computational resources. This apparent "mismatch" is a formidable research challenge.
The research will be carried out in an interdisciplinary environment of several research groups, and under guidance of three supervisors from the Department of Information Security and Communication Technology (IIK), the Department of Computer Science (IDI) and the Department of Computer Science at the United States Naval Academy. The NTNU Internet of Things Lab has experimentation test beds for autonomous sensors and data that can be used as starting point for resource simulations. The Telenor-NTNU AI Lab has expertise within machine learning, decision support and reinforcement learning. The project also benefits from the cross-department research group around Autonomous, Adaptive Sensing (IES, IMF, IIK) and the collaboration with the US Naval Academy provides unique test cases and sources of data, substantial experience with the application of machine learning, and access to massive computational resources.
The appointment is for a term of 3 years without teaching assistance, or up to 4 years including 25% of teaching assistance.
We seek a highly-motivated individual holding a master’s degree in computer science, statistical machine learning, applied mathematics, or communication- and information technology. The average grade must be B or better. A strong background in machine learning and a research-oriented master thesis within one of these areas is expected. Further, the applicants must have experience with programming. Publication activities in the aforementioned disciplines will be considered an advantage. Motivation for fundamental scientific research of practical relevance is essential. The applicant should have good communication skills and be willing to collaborate with other researchers in the cross-disciplinary group of people connected to this project.
See here for regulations concerning appointment and promotion to teaching and research posts: https://www.regjeringen.no/no/dokumenter/regulations-concerning-appointment-and-promotion-to-teaching-and-research-posts/id2519258/
Applicants who do not master a Scandinavian language must provide evidence of good English language skills, written and spoken. The following tests can be used as such documentation: TOEFL, IELTS or Cambridge Certificate in Advanced English (CAE) or Cambridge Certificate of Proficiency in English (CPE). Minimum scores are:
- TOEFL: 600 (paper-based test), 92 (Internet-based test)
- IELTS: 6.5, with no section lower than 5.5 (only Academic IELTS test accepted)
- CAE/CPE: grade B or A.
Appointments are made in accordance with the regulations in force regarding terms of employment for PhD candidates issued by the Ministry of Education and Research, with relevant parts of the additional guidelines for appointment as a PhD candidate at NTNU. Applicants must undertake to participate in an organized PhD programme of study during their period of employment. The person who is appointed must comply with the conditions that apply at any time to employees in the public sector. In addition, a contract will be signed regarding the period of employment.
Applicants must be qualified for admission as PhD students at NTNU. See http://www.ime.ntnu.no/forskning/phd for information about PhD studies at NTNU. Together with the application, include a description of the research work that is planned for completion during the period of the grant.
The position follows code 1017 Research fellow, salary grade 50-62 in the Norwegian State salary scale, gross 436.900 – 537.700 per year, depending on qualifications. A deduction of 2% is made as a statutory contribution to the Norwegian Public Service Pension Fund.
We can offer
- an informal and friendly workplace with dedicated colleagues
- academic challenges
- attractive schemes for housing loan, insurance and pension in the Norwegian Public Service Pension Fund
The Faculty of Information Technology and Electrical Engineering wants to attract outstanding and creative candidates who can contribute to our ongoing research activities. We believe that diversity is important to achieve a good, inclusive working environment. We encourage all qualified candidates to apply, regardless of the gender, disability or cultural background.
For further information about the position, please contact Associate Professor Frank Alexander Kraemer (firstname.lastname@example.org).
The appointment is subject to the conditions in effect at any time for employees in the public sector.
Under Section 25 of the Freedom of Information Act, information about the applicant may be made public even if the applicant has requested not to have his or her name entered on the list of applicants.
The application must be sent electronically through this page ( www.jobbnorge.no) with information about education and relevant experience (all in one combined PDF file). Mark the application with the IE code given below.
The application must contain information of educational background and prior training, exams, and work experience. In addition, the applicant must submit a research statement (max. 3 pages), detailing research interests and initial plans with regard to the above project description. The statement should also describe why the applicant is suited for the position, and how the ROBIOT project relates to previous education, research and competence. Publications and other work that the applicant wishes to be taken into account must be enclosed (including a brief description of the contribution if not obvious).
The application must contain:
- Curriculum vitae (CV) with information about the candidate’s prior training, exams, and work experience
- Certified copies of transcripts and diplomas
- Applicants from universities outside Norway are kindly requested to send a diploma supplement or a similar describes in detail the study and grading system and the rights for further studies associated with the obtained degree
- Research statement (max. 3 pages) including
- A short presentation of the motivation for a PhD study
- How the applicant sees his/her background suitable
- The applicant’s view of research challenges within the area of the PhD position
- How AI or machine learning methods can address the stated problem
- How the competence of the applicant can contribute to solving these challenges
- Names and contact information of at least 2 reference persons
- A copy of the master thesis (in PDF), or, for those who are near to completion of their MSc, an extended abstract combined with a statement of how and when the applicant plans to complete the thesis (1 page, delivery of master thesis no later than June 30th.)
Incomplete applications will not be considered.
Signing of the employment contract should happen within 01.09.2018.
For further information about the application process, please contact Signe J. Talukder (email@example.com).
The application deadline is: April 20, 2018
Mark the application: IE-052-2018
ABOUT THIS JOB
- Deadline Friday, April 20, 2018
- Employer NTNU - Norwegian University of Science and Technology
- Municipality Trondheim
- Place of service Trondheim
- Jobbnorge ID 150404
- Internal ID 2018/9461
- Scope Fulltime
- Duration Temporary
- Applications on this job are registered in an electronic form on jobbnorge.no
- You must complete: Standard CV
- Please refer to where you first saw this job advertised!