PhD Study: Deep-Learning,  Ulster University, UK

Postdoctoral Position in UK United Kingdom

PhD Study: Deep-Learning: The University offers a funded PhD opportunity in the field of Computing, Engineering, and the Built Environment. The project focuses on advancing deep learning methods, particularly Transformer Networks with attention mechanisms, to improve abnormal change detection in various application scenarios such as Earth Observation and anomaly detection in electromagnetic satellite signals.

  • Deep Learning Significance:
    • 🧠 Importance:Deep learning is crucial in machine learning, influencing daily life applications like computer vision, autonomous driving, and earth observation.
    • 🔄 Data-Driven Nature:Deep learning is often data-driven, minimizing the use of domain knowledge in training neural network models.
    • 🛑 Challenges:Traditional deep learning paradigms face challenges in convergence and generalization, especially in machine translation.
  • Transformer Networks:
    • 🌐 Contextual Discrimination:Transformer Networks use attention mechanisms to discern representative data parts based on contextual information, improving model convergence.
    • 🎯 Techniques:Dot-product attention and multi-head attention are common techniques, enhancing the network’s focus on essential data components.
  • Proposed Project:
    • 🔍 Attention Mechanism Development:The project aims to study existing attention techniques and develop new ones for transformer network models.
    • 🌍 Application Scenarios:
      • (1)📸 Image Change Detection:Inspired by human visual attention for abnormal change detection in images.
      • (2)📡 Satellite Signal Anomaly Detection:Focused on contextual characteristics to identify relevant signal portions.
  • Real-world Applications:
    • 🌐 Applicability:The developed transformer networks can find use in monitoring Earth Observation changes, detecting faults in manufacturing processes, and healthcare for human health deterioration.

Essential Criteria:

  • 📚 Academic Requirements:
    • First or Upper Second Class Honours Degree required.
    • Equivalent qualifications, including a Lower Second Class Honours Degree plus a Master’s Degree, may be considered.
    • Professional experience may substitute academic qualifications in exceptional cases.

Desirable Criteria:

  • 🏆 Additional Considerations:
    • First Class Honours (1st) Degree, Masters at 75%, or completion at a commendation/distinction level.
    • Experience in research methods and a strong understanding of the subject area.
    • Publications in peer-reviewed journals are desirable.

Funding and Eligibility:

  • 💰 Scholarship Options:
    • Full, Part, and Fees Only Awards available, covering full-time PhD tuition fees for three years.
    • Research training support grant (RTSG) of * £900 per annum provided.
    • Ineligibility for those with a doctoral degree or more than a year of full-time research.

Department for the Economy (DFE):

  • 🌍 Special Considerations:
    • Home rate tuition fees and a maintenance allowance of * £19,000 per annum for three years.
    • Additional funding for pre-settled/settled EU status and three years’ residency for ROI nationals.

Recommended Reading:

  • 📚 Relevant Publications:
    • List of recommended readings, showcasing the significance of anomaly detection and sequential data analysis in the proposed research area.

Designation: PhD Researcher

Research Area: Development of Change Transformer—Attentive Deep-Learning Approaches for Abnormal Change Detection

Location: Computing, Engineering, and the Built Environment, University (Location not specified)


  • First or Upper Second Class Honours Degree in a relevant subject.
  • Equivalent qualifications may be considered, including a Lower Second Class Honours Degree plus a Master’s Degree with Distinction.
  • Professional experience equivalent to the learning outcomes of an Honours degree may be considered in exceptional circumstances.

Desirable Criteria:

  • First Class Honours (1st) Degree
  • Masters at 75% for VCRS Awards
  • Completion of Masters at a level equivalent to commendation or distinction at Ulster
  • Experience using research methods or approaches relevant to the subject domain
  • Sound understanding of the subject area demonstrated by a comprehensive research proposal
  • Peer-reviewed publications

Funding and eligibility: The University offers Vice Chancellors Research Studentship (VCRS) with the following scholarship options worldwide:

  • Full Award: Full-time tuition fees + £19,000 (tbc)
  • Part Award: Full-time tuition fees + £9,500
  • Fees Only Award: Full-time tuition fees

These scholarships cover full-time PhD tuition fees for three years (subject to academic performance) and provide a £900 per annum research training support grant (RTSG) to support the PhD researcher.

Applicants with a doctoral degree or registered in a full-time doctoral program for more than one year are not eligible.

How to Apply: Submit your CV and a motivation letter (max 2 pages) to the email address specified in the application details by the deadline.

Last Date for Apply: 26 February 2024

Note: Applicants are automatically entered into the competition for the Full Award unless stated otherwise in the application.


Submission Deadline: Monday 26 February 2024 04:00PM

Interview Date: April 2024

Preferred Student Start Date: 16 September 2024


Please enter your comment!
Please enter your name here