Iris Huijben is a Doctoral Candidate in the Signal Processing Systems group of the Electrical Engineering department at Eindhoven University of Technology (TU/e). She received TU/e’s MSc Thesis Award 2020 for her master thesis on sub-Nyquist sampling of medical ultrasound data.
Now she works on representation learning, with applications in sleep medicine, in which she aims to combine signal processing algorithms with deep learning models. By creating meaningful representations of sleep recordings, she intends to learn about the sleeping brain, and therewith bridge the gap between technical models, and clinical interpretation. During the first year of her Doctorate, Iris conducted a research internship regarding deep learning for data compression in the team of Taco Cohen at Qualcomm AI Research, Amsterdam.
Machine learning models can not only be used for automating labor-intensive processes, but also to teach us about the world.
Iris Huijben studied Electrical Engineering at Eindhoven University of Technology (TU/e), where she received her MSc degree cum laude in 2019. Directly after, she joined TU/e as a Doctoral Candidate in the Signal Processing Systems group, where she works on representation learning for human sleep recordings.
A Review of the Gumbel-max Trick and its Extensions for Discrete Stochasticity in Machine LearningIEEE Transactions on Pattern Analysis and Machine Intelligence (2023)
Interpretation and Further Development of the Hypnodensity Representation of Sleep StructurePhysiological Measurement (2023)
Self-Organizing Maps for Contrastive Embeddings of Sleep Recordings(2022)
Certainty about Uncertainty in Sleep StagingSleep (2022)
Representations of temporal sleep dynamicsSleep Medicine Reviews (2022)
- Guest appointment, Onera Health