Qualification Type: | PhD |
---|---|
Location: | Newcastle upon Tyne |
Funding for: | UK Students, EU Students, International Students |
Funding amount: | £18,622 (2023-24 UKRI rate) |
Hours: | Full Time |
Placed On: | 30th January 2024 |
---|---|
Closes: | 1st April 2024 |
Reference: | COMP2147 |
Award Summary
100% Home fees covered and a minimum tax-free annual living allowance of £18,622 (2023-24 UKRI rate)
Overview
We live in a connected world with various smart devices that compose the Internet of Things (IoT). The real breakthrough is processing sensor data streams directly on the low-power devices by the TinyML algorithms before sending them to the computational clouds for further analysis. It enables many always-on battery-operated applications in various domains including smart appliances, smartphones, wearables, smart cars and factories. However, these devices operate in a volatile environment facing various disturbances, so they need to constantly adapt ML models accordingly.
The research aims to develop a new generation of lifelong machine learning algorithms for processing sensor data streams on heterogeneous hardware infrastructures – devices/edge clusters/clouds – to ensure the desired system properties such as security, privacy, and autonomy. The promising approach lies in using computing architectures other than von Neuman, such as neuromorphic devices, customized multi-processors, FPGA, and other dedicated NN coprocessors. Another aspect, no less important, is the ML model security on the devices which ARM Trusted Zones and similar technologies might support. Industry-proven tools such as Tensorflow Lite might not be sufficient for complex heterogeneous IoT environments, and there is a need to develop new algorithms and techniques for model training and deployment.
The research includes designing and evaluating the machine learning algorithms customized for cutting-edge IoT devices and lightweight edge Kubernetes clusters (e.g. MicroK8s, k3s, k0s, Microshift). The candidate will be trained and work with experts within the NUSE research group (with extensive experience in IoT and Machine Learning for embedded devices) and an extensive network of international partners.
Keywords: Machine Learning, TinyML, Edge Computing, Internet of Things
Number Of Awards
1
Start Date
September 2024
Award Duration
3.5 years
Application Closing Date
1 April 2024
Sponsor
Supervisors
Eligibility Criteria
You must have, or expect to gain, a minimum 2:1 Honours degree or international equivalent in a subject relevant to the proposed PhD project which is computer science.
Applicants whose first language is not English require an IELTS score of 6.5 overall with a minimum of 5.5 in all sub-skills.
The studentship covers fees at Home rate (UK and EU applicants with pre-settled/settled status and meet the residency criteria). International applicants are welcome to apply but will be required to cover the difference between Home and International fees.
International applicants may require an ATAS (Academic Technology Approval Scheme) clearance certificate prior to obtaining their visa and to study on this programme.
How To Apply
Apply using Apply to Newcastle Portal
Once registered select ‘Create a Postgraduate Application’.
Use ‘Course Search’ to identify programme of study:
· Search for the ‘Course Title’ using programme code: 8050F
· Research Area: Computing Science
· Select PhD Computer Science as programme of study
You then need to provide the following information in the ‘Further Details’ section:
· A ‘Personal Statement’ (mandatory field) - upload document or write a statement directly in the application form
· The studentship code COMP2147 in the ‘Studentship/Partnership Reference’ field
· When prompted regarding research proposal - select ‘Write Proposal’. You should then type in the title of the research project from this advert you do not need to submit a proposal
Contact Details
Type / Role:
Subject Area(s):
Location(s):