Register now After registration you will be able to apply for this opportunity online.
This opportunity is not published. No applications will be accepted.
Predicting Eye Movements with EEG data
Deep Learning is inspired by the brain structure. But can deep learning help us to advance our understanding of brain functions?
This project aims to develop an eye-tracking approach that offers gaze position that is based on concurrently measured electroencephalography (EEG) data. EEG is a widely-used, safe and cost-friendly method in cognitive neuroscience that directly measures the brain's electrical activity and enables measurement in clinical settings.
Keywords: ML, EEG, Eyetracking, AI glasses
**The Project**
In this project, you will have the opportunity to work on a new large dataset of the simultaneously recorded electroencephalography and eye-tracking data of 450 participants.
We have presented the benchmark for eye movement prediction [1], [2]. Recently, we also introduced a novel framework for time-series segmentation that creates ocular event detectors that do not require additionally recorded eye-tracking modality and rely solely on EEG data [3].
The next project milestone aims to improve the algorithms and understanding of the good performance of the current baselines.
[1] https://openreview.net/forum?id=Nc2uduhU9qa, accepted by NeurIPS 2021,
[2] http://eegeye.net/,
[3] https://arxiv.org/abs/2206.08672
**Your Profile**
- Knowledge in Deep Learning
- Solid background in Machine Learning. Implementation experience with TensorFlow or PyTorch is an advantage
- Willingness to work in an interdisciplinary team
**You would work together with**
- Anh Duong Vo, PhD Candidate in ETH AI Center
- Ard Kastrati, PhD Candidate in Distributed Computing Group, ETH Zurich
- Martyna Plomecka, PhD Candidate at Neuroscience Center Zurich
**The Project** In this project, you will have the opportunity to work on a new large dataset of the simultaneously recorded electroencephalography and eye-tracking data of 450 participants. We have presented the benchmark for eye movement prediction [1], [2]. Recently, we also introduced a novel framework for time-series segmentation that creates ocular event detectors that do not require additionally recorded eye-tracking modality and rely solely on EEG data [3]. The next project milestone aims to improve the algorithms and understanding of the good performance of the current baselines.
[1] https://openreview.net/forum?id=Nc2uduhU9qa, accepted by NeurIPS 2021, [2] http://eegeye.net/, [3] https://arxiv.org/abs/2206.08672
**Your Profile** - Knowledge in Deep Learning - Solid background in Machine Learning. Implementation experience with TensorFlow or PyTorch is an advantage - Willingness to work in an interdisciplinary team
**You would work together with** - Anh Duong Vo, PhD Candidate in ETH AI Center - Ard Kastrati, PhD Candidate in Distributed Computing Group, ETH Zurich - Martyna Plomecka, PhD Candidate at Neuroscience Center Zurich
The goal will be defined together with you :)
The goal will be defined together with you :)
If you are interested, we would love to get to know you! Please contact us:
- Anh Duong Vo, anhduong.vo@ai.ethz.ch
- Martyna Plomecka, martyna.plomecka@uzh.ch
- Ard Kastrati, kard@ethz.ch
If you are interested, we would love to get to know you! Please contact us: - Anh Duong Vo, anhduong.vo@ai.ethz.ch - Martyna Plomecka, martyna.plomecka@uzh.ch - Ard Kastrati, kard@ethz.ch