Engineering PhD programs – 2 PhD Scholarships at Aarhus University, Denmark

    Engineering PhD programs - 2 PhD Scholarships at Aarhus University, Denmark

    2 PhD Scholarships in Deep learning methods, Real-Time IoT Analytics, at Department of Electrical and Computer Engineering, Graduate School of Technical Sciences , Aarhus University, Denmark (Engineering PhD programs)


    Deadline to Apply

    May 01, 2021 (23: 59 GMT +1)


    Overview

    PositionPhD Scholarships
    No. of Position(s)2
    Research Area– Deep learning methods
    – Real-Time IoT Analytics
    ScholarshipAccording to Standard Norms
    WorkplaceDepartment of Electrical and Computer Engineering
    Finlandsgade 22, 8200 Aarhus N
    Denmark
    Contract PeriodNot specified

    Projects

    Using deep learning methods to tailor sleep scoring algorithms to specific populations

    Real-Time IoT Analytics at Edge  

    How to Apply?

    To access the application form, click the apply now link above

    Fill in the following information

    • Personal information
    • Academic background
    • Admission
    • Financing (if any)
    • Study: In the dropdown menu you must choose the above-mentioned project you interested
      • for example “Using deep learning methods to tailor sleep scoring algorithms to specific populations
    • Source (how you found out about the call)
      • Please mention that “nViews Career”

    Next to some of the information fields you will find a number. Click on the number to get further directions on how to fill in the information field/what information is needed.

    Documents Required

    • One reference (template for references)
    • Curriculum vitae,
    • Motivation (max. 1 page)
    • Transcripts, grade point averages (weighted and unweighted), and diploma(s) for both Bachelor’s and Master’s degree. If the original documents are not in English or one of the Scandinavian languages (i.e. Norwegian, Swedish or Danish) then copies of the original documents as well as a certified English translation must be attached.
    • Project description (½-4 pages). For technical reasons, you must upload a project description. When – as here – you apply for a specific project, please simply copy the project description above, and upload it as a PDF in the application. If you wish to, you can add to this description and you can indicate an URL where further information can be found. Please note that we reserve the right to remove scientific papers, large reports, theses and the like.
    • Documentation of language skills if required.

    Documentation of language skills

    The English language requirement at the graduate school is comparable to an “English B level” in the Danish upper secondary school (“gymnasium”).

    English language qualifications comparable to an “English B level” are documented by one of the following tests:

    • TOEFL test (internet-based), minimum score: 83. The graduate school does not accept the paper-based test, nor the TOEFL ITP test. Remember to ask the test center to send your test results to Aarhus University in order to enable verification of your test results. Aarhus University’s TOEFL code is 8935.
      Currently, the TOEFL iBT Special Home Edition test (available in selected areas) will also be accepted.
    • IELTS (academic) test, minimum average score: 6.5 points
    • Cambridge English Language Assessment:
      Cambridge Certificate of Proficiency (CPE)
      Cambridge English: Certificate of Advanced English with grade A,B or C (CAE)
      Cambridge English: First Certificate with grade A (FCE)

    When to take the test and how to upload the documentation:
    The test result must not be more than two years old at the time of application.

    The English language test should be taken before applying for admission and uploaded under “language skills documentation” in the online application form.

    Note

    • The programme committee may request further information or invite the applicant to attend an interview.
    • The project will only be initiated if final funding (from the graduate school/the faculty) is secured.

    Using deep learning methods to tailor sleep scoring algorithms to specific populations

    Qualifications and specific competences

    The successful applicant should have a strong quantitative background (i.e., bachelor’s or master’s degree in engineering, physics, mathematics or similar), and an interest in either machine learning, neuroscience, brain computer interfaces or biomedical engineering in general. The project is expected to entail a great deal of work with conventional deep learning libraries (tensorflow, pytorch or similar), so experience with those and/or scientific programming tools (matlab, python, R, C/C++) is a definite advantage.

    Given the explorative nature of scientific research, an interest and willingness to learn new skills is an important quality – perhaps even more important than the specific skillset held by the applicant at the time of application.

    Our group works in a collaborative and explorative fashion. This means that good communication skills, both written and oral, are important to carrying out the everyday work.

    At the time of writing, the group consists of 1 full professor, 2 assistant professors, 2 post docs and 3 phd students, spanning 5 different nationalities.

    Research area and project description

    In recent years, great progress has been made in developing algorithms for automatic sleep scoring. This is a societal good, because low cost sleep scoring will benefit many different branches of health care.

    However, sleep scoring is also an interesting ‘model problem’ for developing new methods in signal analysis. Large data bases for training and testing exist, and the problem is both complex and important. An exciting challenge is the fact that training data is not equally distributed among patient groups, and some groups, in particular those that are hardest to score, have relatively little training data.

    In this project, the candidate will develop and test approaches to ‘transfer learning’ or ‘semi supervised learning’, in which high performing algorithms trained on large data sets will be transformed to perform similarly well on smaller data sets for rarer types of sleep recordings.

    The developed methods can be used for regular, clinical sleep recordings, but an important task will also be in testing the methods for data recorded using the “ear-EEG” platform, which has been developed in our group. In general, the methods developed in this project will likely be relevant for most mobile sleep monitoring platforms, of which there are a growing number.

    Contacts:

    Applicants seeking further information are invited to contact:

    Kaare Mikkelse
    Email: mikkelsen.kaare@ece.au.dk


    Real-Time IoT Analytics at Edge

    Qualifications and specific competences

    We are looking for highly motivated and independent students willing to take the challenge to do a successful 3-year PhD programme in Aarhus University. The ideal candidate will have the following profile (but not all items are required for a successful application):

    • Relevant Master’s degree (e.g., Computer Engineering, Computer Science, Software Engineering, Electrical Engineering), although exceptional candidates from related disciplines (e.g., Applied Mathematics) will also be considered.
    • Excellent undergraduate and master degree grades are required.
    • Background on data analytics, machine learning, information theory, data storage is highly desired, but candidates from other disciplines will be considered based on their merits and potential.
    • Background on linear algebra, mathematics and statistics is desired.
    • Strong programming skills in Python or C++.
    • Good English verbal and written skills are required.

    Research area and project description

    The proliferation of the Internet of Things (IoT) and the ever-increasing massive IoT data provide unprecedented opportunities for innovations. How to extract value from the massive IoT data has gained significant interest among researchers and industry. Conventionally, historical analytics is often used to obtain insights from the mining of historical data for diagnostic and descriptive purposes. On the other hand, real-time IoT analytics promises to realize proactive and predictive analytics by analyzing IoT data as soon as it enters the system in a predefined timeframe; that is becoming a new trend in IoT data analytics and has applications in diverse verticals, such as smart home, industrial IoT, smart grid, E-health, smart transportation and many others. Real-time IoT analytics at the network edge would significantly reduce the analytics response time and save the bandwidth to forward all the data to the cloud. However, the analytics capability of edge computing is not as powerful as that of cloud computing. Therefore, the question is not how to perform analytics on massive IoT data, but rather how to perform analytics on the right data.

    In this project, we will develop an edge analytics framework for real-time IoT data analytics to address the limitations of existing data-center-based analytics:

    1. We will study existing software platforms for data analytics so as to: (i) examine what extent they support real-time IoT data analytics at Edge, and (ii) understand their performance tradeoffs and deficiencies.
    2. We will develop tailor-made techniques for real-time IoT stream data analytics at Edge, leveraging our previous expertise in data engineering.
    3. We will study the relationship between sensor data representation, data storage architecture, and data analytics, to understand their impact on latency, accuracy, scalability, and fault tolerance.
    4. We will optimize the sensor data representation and compression not only for transmission but also for facilitating and accelerating data analytics.
    5. We will develop network methodologies to facilitate edge and cloud collaboration for real-time data analytics.

    Contacts:

    Applicants seeking further information are invited to contact:

    Assoc. Prof. Qi Zhang
    E-mail: qz@ece.au.dk

    Assoc. Prof. Panagiotis Karras
    Email: panos@cs.au.dk

    Official advertisement

    PhD fellow in Traceable Pharmaceuticals

    Other PhD Positions with Scholarships

    Post expires at 8:59am on Sunday May 2nd, 2021 (GMT+9)