PhD-TA in Robust Machine Learning for Dynamic and Out-of-Distribution Data

Are you passionate about exploring robust machine learning models for temporal data? Join our research team to explore cutting-edge methods that improve predictive models' robustness to distribution shifts, data noise, and other real-world challenges.

Information
Are you excited to tackle challenges in machine learning and data mining? In this PhD position, you will explore cutting-edge machine learning methods aimed at improving the robustness of predictive models, with a particular focus on temporal data—data that captures how systems evolve over time.

Robustness is essential for deploying machine learning models in real-world settings where data distributions can shift unexpectedly due to changes in environment, behavior, or context. Models that lack robustness may perform well in controlled settings but fail in practice, especially when exposed to Out-of-Distribution (OOD) data. Ensuring reliable performance in such scenarios is key for building systems that can be trusted in high-stakes domains like healthcare, social platforms, and e-commerce.

Your research will tackle several key challenges, including but not limited:

• Out-of-Distribution (OOD) Generalization: You will investigate how to design models that generalize well to unseen distributions, such as domain shifts and temporal drifts. This includes studying algorithms that models domain-invariant representation learning, adaptive modeling, or causaliy driven approaches.
• Multi-modal Modeling: Real-world systems often involve multiple data sources (e.g., text, time series, sensor data). You will explore how to understand, integrate and model this multi-modal information to enhance prediction accuracy and model resilience.
• Event Sequence Modeling: Many human-centered applications involve irregular, context-dependent event sequences (e.g., protests, clinical visits) that are sensitive to distribution shifts. You will develop explainable models for predicting human event occurrences and ensuring transferability to unseen social scenarios, studying advances in sequential deep learning and large language models (LLMs).

You will work with experts from machine learning and data mining. You work will contribute to the Data Mining Group in Data and AI cluster.

Do you recognize yourself in this profile and would you like to know more? Please contact the hiring manager Amy Deng, Assistant Professor, Department of Mathematics and Computer Science, s.deng@tue.nl or Mykola Pechenizkiy, Professor, Department of Mathematics and Computer Science, m.pechenizkiy@tue.nl.

Visit our website for more information about the application process or the conditions of employment. You can also contact HRServices.MCS@tue.nl.

Are you inspired and would like to know more about working at TU/e? Please visit our career page.