Federated Unlearning for Neuro-Symbolic Autonomous Driving

About the Project

Federated learning (FL) enables vehicles to collaborate on model training without sharing raw data, while neuro-symbolic AI combines perception from neural networks with rule-based reasoning for compliance and explainability. Yet when data must be erased—because it is biased, malicious, or withdrawn—current federated unlearning techniques are computationally heavy, weakly verified, and vulnerable to attacks. This PhD tackles those gaps for safety-critical autonomous driving.

Aim

Design, implement, and validate a federated unlearning framework and algorithms for neuro-symbolic autonomous driving models that (i) efficiently remove data influence, (ii) preserve symbolic logic, (iii) provide reliable verification, and (iv) remain robust to adversarial threats while supporting regulatory compliance.

Objectives

1. Create an evaluation framework spanning adversarial and non-adversarial settings with diverse metrics.
2. Develop a scalable unlearning method tailored to neuro-symbolic models.
3. Quantify trade-offs among accuracy, efficiency, privacy, and robustness under client heterogeneity.
4. Ensure procedures align with GDPR-style “right to erasure” and produce audit-ready evidence.

Expected Outputs

1. A reusable framework to assess effectiveness and robustness of federated unlearning.
2. A novel unlearning approach addressing neuro-symbolic challenges in autonomy.
3. An adversarial testbed for rigorous robustness evaluation.
4. Best-practice guidelines for compliant, trustworthy deployments.

Training

The candidate will develop expertise in federated learning, unlearning theory and systems, neuro-symbolic AI, security evaluation, and reproducible ML engineering, positioning them for leadership in trustworthy autonomy.

Academic qualifications

First degree (minimum 2:1 classification) in Computer Science, Electronic Engineering, Machine Learning, Artificial Intelligence

English language requirement

IELTS score must be at least 6.5 (with not less than 6.0 in each of the four components). Other, equivalent qualifications will be accepted. Full details of the University’s policy are available online.

Essential attributes:

• Fundamental knowledge in the following areas: Federated learning and distributed optimisation, Computer vision & perception for autonomy; sensor fusion and time-series modelling and Unlearning theory & auditing
• Experience in fundamental cybersecurity or system security
• Competent in programming and critical analysis
• Knowledge of machine learning
• Good written and oral communication skills
• Strong motivation, with evidence of independent research skills relevant to the project
• Good time management

Desirable attributes:

• Federated/distributed learning concepts: client heterogeneity, aggregation, communication/compute trade-offs.
• Security & robustness literacy: adversarial/threat models (poisoning, backdoor, membership inference) and empirical evaluation.
• Neuro-symbolic reasoning appetite: willingness to engage with logic constraints (e.g., rules/constraints that must hold post-unlearning)

APPLICATION CHECKLIST

• Completed application form
• CV
• 2 academic references, using the Postgraduate Educational Reference Form (download)
• Research project outline of 2 pages (list of references excluded). The outline may provide details about
  1. Background and motivation of the project. The motivation, explaining the importance of the project, should be supported also by relevant literature. You can also discuss the applications you expect for the project results.
  2. Research questions or objectives.
  3. Methodology: types of data to be used, approach to data collection, and data analysis methods.
  4. List of references.
• The outline must be created solely by the applicant. Supervisors can only offer general discussions about the project idea without providing any additional support.
• Statement no longer than 1 page describing your motivations and fit with the project.
• Evidence of proficiency in English (if appropriate)

To be considered, the application must use

• the advertised title as project title

For informal enquiries about this PhD project, please contact z.tan@napier.ac.uk

PhD Start Date: October 2026

Funding Notes

International applicants should note that visa application costs and the NHS health surcharge are additional costs to be taken into consideration, and successful applicants will need to cover these expenses themselves.