PhD on Trustworthy and Secure AI with Focus on OOD Detection
Updated: 09 Dec 2024
We are seeking an enthusiastic PhD candidate to develop novel ideas to establish trust in deep learning models. Trustworthy AI is a major topic in machine learning, which is illustrated by the increasing number of initiatives to enforce AI systems to be more trustworthy. Although machine learning models typically perform well on input data they are trained on, they are less suited for indicating that they cannot provide a reliable prediction as the input is too different from the data they are trained on.
Our goal is to provide theoretically founded Out-Of-Distribution (OOD) methods as a stepping stone toward trustworthy machine learning models that know what they don't know. We can distinguish different types of OOD data, such as adversarial examples, near-OOD, and far-OOD. A challenge here is that, unlike for adversarial examples, we don't have a mathematical framework (yet) for near-OOD and far-OOD data. The topic of OOD detection also links to explainable AI, whose exploration might be worthwhile to identify differences in the processing of in-distribution data and OOD data.
This project is a collaboration between the Data and AI cluster from TU/e and the industrial semiconductor company NXP. This gives the position a unique opportunity to experience and build a network in both academic industrial research.
The position is available from October 1st and will be performed under the supervision of Profs. Wil Michiels (Security Group, NXP), and Sibylle Hess (Data and AI).
Requirements:
- A master's degree (or an equivalent university degree) in Mathematics, Computer Science, or a related field.
- Experience in programming and empirical analysis in Deep Learning (e.g. in Python, PyTorch).
- Excellent problem-solving skills and ability to work independently and collaboratively.
- Strong written and oral communication skills in English.
Salary Benefits:
A meaningful job in a dynamic and ambitious university, in an interdisciplinary setting and within an international network. You will work on a beautiful, green campus within walking distance of the central train station. In addition, we offer you:
- Full-time employment for four years, with an intermediate evaluation (go/no-go) after nine months. You will spend 10% of your employment on teaching tasks.
- Salary and benefits (such as a pension scheme, paid pregnancy and maternity leave, partially paid parental leave) in accordance with the Collective Labour Agreement for Dutch Universities, scale P (min. €2,872 max. €3,670).
- A year-end bonus of 8.3% and annual vacation pay of 8%.
- High-quality training programs and other support to grow into a self-aware, autonomous scientific researcher. At TU/e we challenge you to take charge of your own learning process.
- An excellent technical infrastructure, on-campus children's day care and sports facilities.
- An allowance for commuting, working from home and internet costs.
- A Staff Immigration Team and a tax compensation scheme (the 30% facility) for international candidates.
38 hours per week
De Rondom 70