The
Decentralized Artificial Intelligence Research Lab (DARL) at the Eindhoven University of Technology is seeking 2 talented and passionate Ph.D. candidates to join our team. These positions are part of the
AiNed Fellowship project 'Private Ears, Shared Insights' funded by the
NWO. Our mission is to revolutionize the field of Artificial Intelligence (AI) by developing cutting-edge collaborative learning techniques that enable AI models to learn from large-scale decentralized data while preserving user privacy. Our ultimate goal is to instill self-learning capabilities in globally distributed computational devices for everyday use.
The field of AI has seen unprecedented advancements in recent years, driven by the development of foundation models that have expanded the boundaries of machine capabilities. However, learning these models requires direct access to vast data repositories, which poses significant privacy and logistical challenges, especially in the health sensing domain that involves personal data. To address this, the DARL is at the forefront of research exploring decentralized and collaborative approaches to developing unified AI systems. Our research entails the development of novel methodologies at the intersection of self-supervised learning, data-centric machine learning, trustworthy AI, and human-machine collaboration (i.e., expert-in-the-loop) for healthcare and high-tech industries.
We are currently seeking candidates for two PhD positions, each focusing on a cutting-edge research topic within the field of audio (and speech) understanding for health monitoring:
1. Federated Audio Foundation Models: Advance pre-training of foundation models with unlabeled private data that bring label efficiency in data-constrained environments, few-shot emerging capabilities, rapid adaptation, and synthetic data generation.
2. Domain Knowledge-Augmented Representation Learning: Development of techniques for incorporating clinical domain knowledge with physics-informed neural networks, weak supervision, active learning, and/or cross-modal knowledge transfer to improve data efficiency, generalizability, and justifiability of deep models.
These positions provide an opportunity to advance the frontier of decentralized machine learning and distributed sensing systems. Successful candidates will have the opportunity to work with a dynamic and interdisciplinary team of researchers, collaborating with experts in AI, healthcare, and industry.