UNIVERSITY OF COPENHAGEN

Machine learning algorithms are increasingly being deployed in real-world systems that utilize user data for training or interact directly with individuals—often making decisions that have significant impacts on people’s lives. In this context, ensuring the trustworthiness of these systems, particularly concerning privacy, unlearning, and/or robustness, is important. We are seeking a highly motivated postdoctoral researcher to conduct theoretical and/or empirical research on privacy and/or robustness in machine learning.

Potential research topics include, but are not limited to:

• Learning-theoretic implications of differential privacy, machine unlearning, or robustness and the relationship between these notions
• Developing statistically and computationally efficient algorithms that incorporate these properties
• Exploring privacy, unlearning, and/or robustness in alternative learning settings such as online, active, semi-supervised, or noisy environments

We highly encourage applicants with fresh ideas and interests beyond these specific topics, but related to the broader description of privacy and robustness, to apply.
The successful candidate will be based at the Department of Computer Science at the University of Copenhagen and will work primarily with Amartya Sanyal. Their main duty will be to conduct novel research in the aforementioned areas, collaborate with other members in the research group, and publish findings in leading venues such as NeurIPS, COLT, ALT, ICML, SODA, FOCS, AISTATS, UAI, SaTML, and ICLR. Additional responsibilities may include collaborating with other research groups internationally and domestically as well as participating in other related duties as needed. Inquiries about the position can be made to Amartya Sanyal (amsa@di.ku.dk).
The position is open from June, 2025 or as soon as possible thereafter. Earlier start date can be discussed. The length of the employment is 2 years.
Further information on the Department can be found at https://di.ku.dk/english/.
The University wishes our staff to reflect the diversity of society and thus welcomes applications from all qualified candidates regardless of personal background.

POSITION REQUIREMENTS:

• (Required) PhD degree in computer science, statistics, mathematics, or similar
• (Required) Experience with some aspects of learning theory, differential privacy, or robustness.
• (Required) Basic Programming skills in Python.
• (Required) High level of motivation with the ability to conduct independent research by identifying key problems and driving projects to completion.
• (Required) Proven skills in writing high-quality research papers and delivering effective presentations.
• (Optional) Experience in empirical privacy and robustness, training of machine learning models, or the desire to get involved in it.
• (Optional) Familiarity with some aspects of machine unlearning.

Please refer the Job description for details