PhD Studentship: Neuromorphic Photonic Processor for Scalable Generative AI Applications

Project title:

Neuromorphic Photonic Processor for Scalable Generative AI Applications

Research Group:

George Green Institute for Electromagnetics Research, Faculty of Engineering,
University of Nottingham, UK

Project description

The research topic focuses on developing energy-efficient and high-speed solutions for generative AI using neuromorphic photonics. Specifically, it addresses the challenge of handling the nonlinear, probabilistic operations that generative AI tasks demand, which traditional digital computing systems struggle to process. Open problems in neuromorphic photonic reservoir computing, such as stability, scalability, and memory integration, will also be explored.

The project:

The candidate is expected to develop a photonic reservoir computing system on a silicon chip platform, harnessing chaotic nonlinear dynamics to perform computations. This involves developing analytical and numerical models to predict the impact of external perturbations on system performance. The candidate will prototype and characterise the photonic chip system to demonstrate its ability to efficiently process generative AI tasks.

This PhD vacancy is open to UK (home), EU, and international students.

The candidate will have access to the GGIEMR Institute’s supercomputer and dedicated laboratories for the fabrication and testing of optical fibres, chips, and systems, including a class 10,000 cleanroom. Available facilities include a comprehensive laser suite, a dedicated optical bench for optical waveguide and component characterisation, photoluminescence equipment, Fourier-transform infrared (FTIR) spectroscopy, refractive index measurement tools, and near-field imaging systems.

The ideal candidate will have:

1. A first or upper second-class honours or Master's degree in Physics, Applied Physics, Electrical and Electronic Engineering, Mathematical Sciences, or a related subject from a recognised institution.
2. A solid background in design and/or experimental photonics or optoelectronics. Previous experience in using photonic design software (Lumerical, Comsol, MEEP or HFSS) will be an advantage.
3. A solid understanding of electromagnetics, mathematics and statistics, and machine learning theory/algorithms, with excellent analytical, numerical, and problem-solving skills.
4. Strong programming skills in Matlab, C/C++, or Python. Hands-on experience with deep learning platforms (e.g., TensorFlow, PyTorch) is advantageous.
5. Experience with cleanroom processes, photonic chip fabrication, and optical characterisation tools.
6. Excellent scientific writing, presentation, and communication skills, with fluency in English.
7. A proactive, independent, and self-reliant work attitude, with the ability to thrive in a fast-paced and collaborative research environment.

If English is not your first language, you must provide evidence that you meet the University’s minimum English Language requirements.

How to apply

Interested candidates should follow a two-stage application process:

Stage 1: Expression of Interest and Interview

1. Submit the following documents directly to Dr Sendy Phang ([email protected]) by 7 February 2025
   1. CV
   2. Cover letter explaining your research interests, relevant skills and experience, and why you are interested in this PhD project
   3. Academic transcripts (for both undergraduate and postgraduate degrees, if applicable)
   4. Copies of any publications or an example of your technical writing, such as a project report or dissertation
2. Shortlisted candidates will be invited for an online interview to assess their suitability for the project on 10 February 2025. Successful candidates will then proceed to the next stage.

Stage 2: Formal PhD Study Application

1. Candidates who pass the interview stage will be advised to formally apply for PhD admission at https://www.nottingham.ac.uk/pgstudy/how-to-apply/apply-online.aspx
2. Once admitted, I will submit an application for internal funding on behalf of the successful candidate. Please note that securing admission to the PhD program does not automatically guarantee funding.

Contact details

Name: Dr Sendy Phang

Email: [email protected]

Closing Date: 07 Feb 2025
Category: Studentships