PhD Candidate in AI and Hardware Security

University of Amsterdam; Netherlands; On-site; EUR 36,000 - 45,000; Full time

Job summary

University of Amsterdam's Parallel Computing Systems group seeks a highly motivated candidate with a computer security and/or machine learning background to study vulnerabilities of AI systems and to develop robust defenses.

You will design and perform security evaluations on AI algorithms and hardware, publish your results at leading venues, and complete a four-year PhD with teaching duties and international collaborations.

Qualifications

• MSc in computer science or related field.
• Strong background in AI and hardware security.
• Demonstrated research ability (publications or thesis).
• Willingness to collaborate with researchers.
• Professional command of English and presentation skills.

Responsibilities

• Design and carry out security evaluations on AI algorithms and hardware, incl. model extraction, fault injection and side-channel analysis.
• Investigate how weaknesses arise across algorithm, software and hardware layers and develop models to explain them.
• Develop hardware-aware countermeasures that are robust and low-overhead.
• Validate methods experimentally on real hardware in the lab, not only in simulation.
• Publish results at leading conferences/journals and complete a PhD within four years.
• Present work at international venues and build collaborations with academia and industry.
• Contribute to teaching (about 10% of time) as TA or supervising students.

Skills

AI security

Hardware security

Research ability

English communication

Education

MSc in computer science or related field

Job description

We are looking for a highly motivated candidate with a computer security and/or machine learning background to join the Parallel Computing Systems (PCS) group at the University of Amsterdam. We look for vulnerabilities of AI systems across the whole AI system: from adversarial examples, backdoors and model extraction attacks to physical side-channel and fault injection attacks on AI chips. We build defenses that are both robust and efficient. A strong background in machine learning, hardware or security, together with the drive to develop expertise in the other areas, is highly valued. You'll get hands-on access to the hardware lab and close supervision.

This is what you will do

In this project, you will work to secure AI hardware by investigating vulnerabilities at the algorithm, software, and hardware levels and developing countermeasures to address them. You will study how weaknesses in one layer can be exploited through another. For example, side-channel leakage from an accelerator can be used to extract neural network parameters.

Tasks and responsibilities

• designing and carrying out realistic security evaluations on AI algorithms and hardware, such as model extraction, fault injection, and side-channel analysis, on platforms including NPUs, GPUs and other AI chips;
• investigating how vulnerabilities arise and how weaknesses at the algorithm, software, and hardware levels interact, and developing models and tools to explain these mechanisms;
• developing models and tools to explain why these vulnerabilities arise. Designing hardware-aware countermeasures that are robust, low-overhead and practical to deploy;
• validating your methods experimentally on real device in hardware lab, not only in simulation;
• publishing your results in leading international conference and journals, and completing a PhD thesis within four years;
• presenting your work at international venues and helping to build collaborations with academic and industrial partners;
• contributing to teaching (about 10% of your time), for example as a teaching assistant or by supervising Bachelor and Master students.

What we ask of you

Your experience and profile

• an MSc in computer science or a related field;
• a strong background and interest in AI and hardware security;
• a demonstrated ability to perform research (for instance in the form of publications or a Master thesis);
• the willingness to work collaboratively with other researchers;
• professional command of English and presentation skills

This is what we offer you

A temporary contract for 38 hours per week for the duration of 4 years (the initial contract will be for a period of 18 months and after satisfactory evaluation it will be extended for a total duration of 4 years).The preferred starting date is as soon as possible.This should lead to a dissertation (PhD thesis). We will draft an educational plan that includes attendance of courses and (international) meetings. We also expect you to assist in teaching undergraduates and master students.

The gross monthly salary, based on 38 hours per week and dependent on relevant experience, ranges between € 3,204 to € 4,051 (scale P). This does not include 8% holiday allowance and 8,3% year-end allowance. The UFO profile PhD Candidate is applicable. A favourable tax agreement, the ‘30% ruling’, may apply to non-Dutch applicants. The Collective Labour Agreement of Universities of the Netherlands is applicable.

Curious about our extensive secondary benefits package? You can read more about it here.

You will work in this team

The Faculty of Science has a student body of around 8,000, as well as 1,800 members of staff working in education, research or support services. Researchers and students at the Faculty of Science are fascinated by every aspect of how the world works, be it elementary particles, the birth of the universe or the functioning of the brain.

The mission of the Informatics Institute (IvI) is to perform curiosity-driven and use-inspired fundamental research in Computer Science. The main research themes are Artificial Intelligence, Computational Science and Systems and Network Engineering. Our research involves complex information systems at large, with a focus on collaborative, data driven, computational and intelligent systems, all with a strong interactive component.

The Parallel Computing Systems (PCS) group at the University of Amsterdam performs research on the design, programming and run-time management of parallel and distributed computer systems. The modeling, analysis and optimization of the extra-functional aspects of these systems, such as performance, power/energy consumption, thermals, reliability, and security but also the degree of productivity to design and program these systems, play a pivotal role in this work.

Want to know more about our organisation? Read more about working at the University of Amsterdam.

If you have any questions or do you require additional information? Please contact:

• Zhuoran Liu, Assistant Professor, [e-mail]