Ivo Pascal de Jong

Abstract: Public Motor Imagery-based brain-computer interface (BCI) datasets are being used to develop increasingly good classifiers. However, they usually follow discrete paradigms where participants perform Motor Imagery at regularly timed intervals. It is often unclear what changes may happen in the EEG patterns when users attempt to perform a control task with such a BCI. This may lead to generalisation errors. We demonstrate a new paradigm containing a standard calibration session and a novel BCI control session based on EMG. This allows us to observe similarities in sensorimotor rhythms, and observe the additional preparation effects introduced by the control paradigm. In the Movement Related Cortical Potentials we found large differences between the calibration and control sessions. We demonstrate a CSP-based Machine Learning model trained on the calibration data that can make surprisingly good predictions on the BCI-controlled driving data. Open Access Link

2024

Uncertainty Quantification for Cross-Subject Motor Imagery Classification

9th Graz BCI Conference. Prithviraj Manivannan, Ivo Pascal de Jong, Matias Valdenegro-Toro, Andreea Ioana Sburlea

Abstract: Uncertainty Quantification aims to determine when the prediction from a Machine Learning model is likely to be wrong. Computer Vision research has explored methods for determining epistemic uncertainty (also known as model uncertainty), which should correspond with generalisation error. These methods theoretically allow to predict misclassifications due to inter-subject variability. We applied a variety of Uncertainty Quantification methods to predict misclassifications for a Motor Imagery Brain Computer Interface. Deep Ensembles performed best, both in terms of classification performance and cross-subject Uncertainty Quantification performance. However, we found that standard CNNs with Softmax output performed better than some of the more advanced methods. Open Access Link

2024

How disentangled are your classification uncertainties?

arXiv preprint. Ivo Pascal de Jong, Andreea Ioana Sburlea, Matias Valdenegro-Toro

Abstract: Uncertainty Quantification in Machine Learning has progressed to predicting the source of uncertainty in a prediction: Uncertainty from stochasticity in the data (aleatoric), or uncertainty from limitations of the model (epistemic). Generally, each uncertainty is evaluated in isolation, but this obscures the fact that they are often not truly disentangled. This work proposes a set of experiments to evaluate disentanglement of aleatoric and epistemic uncertainty, and uses these methods to compare two competing formulations for disentanglement (the Information Theoretic approach, and the Gaussian Logits approach). The results suggest that the Information Theoretic approach gives better disentanglement, but that either predicted source of uncertainty is still largely contaminated by the other for both methods. We conclude that with the current methods for disentangling, aleatoric and epistemic uncertainty are not reliably separated, and we provide a clear set of experimental criteria that good uncertainty disentanglement should follow. Open Access Link

2023

Uncertainty Quantification in Machine Learning for Biosignal Applications - A Review

arXiv preprint. Ivo Pascal de Jong, Andreea Ioana Sburlea, Matias Valdenegro-Toro

Abstract: Uncertainty Quantification (UQ) has gained traction in an attempt to fix the black-box nature of Deep Learning. Specifically (medical) biosignals such as electroencephalography (EEG), electrocardiography (ECG), electroocculography (EOG) and electromyography (EMG) could benefit from good UQ, since these suffer from a poor signal to noise ratio, and good human interpretability is pivotal for medical applications and Brain Computer Interfaces. In this paper, we review the state of the art at the intersection of Uncertainty Quantification and Biosignal with Machine Learning. We present various methods, shortcomings, uncertainty measures and theoretical frameworks that currently exist in this application domain. Overall it can be concluded that promising UQ methods are available, but that research is needed on how people and systems may interact with an uncertainty model in a (clinical) environment. Open Access Link

Research

I'm interested in how Machine Learning methods behave when subjected to the reality of datasets that are often noisy, sparse or otherwise non-ideal. The research I do on Uncertainty in Machine Learning is therefore largely focused on empirical experiments where data's and algorithms interact. I'm specifically interested in distinguishing between data uncertainty and model uncertainty in Deep Learning.
For Brain-Computer Interfaces I'm interested in applying these Uncertain Machine Learning methods and seeing whether they give meaningful benefits to researcher or users. Personally, I'm most interested in Brain-Computer Interfaces when they can be useful for ALS, spinal cord injury, or stroke patients. Through decoding brain signals from movement attempts, I aim to achieve meaningful and useable control of a device.
I've selected some highlighted papers on the left. All publications are freely available as Open Access, just click the link.

Collaboration & Supervision

Interesting research comes from sharing interesting ideas, and many of my papers come from supervising excellent students or collaborating with excellent researchers. I am currently open to collaborations with:

Bachelor or Master students looking to do a research project.
Businesses interested in internship projects or with an interesting research problem.
Researchers looking to work with (uncertain) Machine Learning or Brain-Computer Interfaces.
Educators interested in AI.

If you're considering working with me, feel free to get in touch. I'd be happy to have a cup of coffee with you!

Get in touch!

Lecturer & PhD student

Research

Collaboration & Supervision

Teaching

Publicity & Outreach