2000 character limit reached
Uncertainty Quantification for cross-subject Motor Imagery classification (2403.09228v1)
Published 14 Mar 2024 in cs.LG and cs.CV
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.
- “MOABB: trustworthy algorithm benchmarking for BCIs” In Journal of neural engineering 15.6 IOP Publishing, 2018, pp. 066011
- “A review of uncertainty quantification in deep learning: Techniques, applications and challenges” In Information Fusion 76, 2021, pp. 243–297 DOI: https://doi.org/10.1016/j.inffus.2021.05.008
- “UNCER: A framework for uncertainty estimation and reduction in neural decoding of EEG signals” In Neurocomputing 538 Elsevier, 2023, pp. 126210
- “Monte Carlo Dropout for Uncertainty Estimation and Motor Imagery Classification” In Sensors 21.21, 2021 DOI: 10.3390/s21217241
- Egor I Chetkin, Sergei L Shishkin and Bogdan L Kozyrskiy “Bayesian opportunities for brain–computer interfaces: Enhancement of the existing classification algorithms and out-of-domain detection” In Algorithms 16.9 MDPI, 2023, pp. 429
- “Robust Motor Imagery Tasks Classification Approach Using Bayesian Neural Network” In Sensors 23.2 MDPI, 2023, pp. 703
- “Robust and efficient uncertainty aware biosignal classification via early exit ensembles” In ICASSP 2022-2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2022, pp. 3998–4002 IEEE
- “On Calibration of Modern Neural Networks” In Proceedings of the 34th International Conference on Machine Learning 70, Proceedings of Machine Learning Research PMLR, 2017, pp. 1321–1330
- Simon JD Prince “Computer vision: models, learning, and inference” Cambridge University Press, 2012
- Ivo Pascal de Jong, Andreea Ioana Sburlea and Matias Valdenegro-Toro “Uncertainty Quantification in Machine Learning for Biosignal Applications–A Review” In arXiv preprint arXiv:2312.09454, 2023
- “Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learning”, 2016 arXiv:1506.02142 [stat.ML]
- “DropConnect Is Effective in Modeling Uncertainty of Bayesian Deep Networks”, 2019 arXiv:1906.04569 [cs.LG]
- Balaji Lakshminarayanan, Alexander Pritzel and Charles Blundell “Simple and scalable predictive uncertainty estimation using deep ensembles” In Advances in neural information processing systems 30, 2017
- “Flipout: Efficient pseudo-independent weight perturbations on mini-batches” In arXiv preprint arXiv:1803.04386, 2018
- “Uncertainty estimation using a single deep deterministic neural network” In International conference on machine learning, 2020, pp. 9690–9700 PMLR
- “Review of the BCI Competition IV” In Frontiers in Neuroscience 6, 2012 DOI: 10.3389/fnins.2012.00055
- “Deep learning with convolutional neural networks for EEG decoding and visualization” In Human Brain Mapping, 2017 DOI: 10.1002/hbm.23730
- “MEG and EEG data analysis with MNE-Python” In Frontiers in Neuroscience 7, 2013, pp. 267 DOI: 10.3389/fnins.2013.00267
- Navneet Tibrewal, Nikki Leeuwis and Maryam Alimardani “The Promise of Deep Learning for BCIs: Classification of Motor Imagery EEG using Convolutional Neural Network” In bioRxiv Cold Spring Harbor Laboratory, 2021 DOI: 10.1101/2021.06.18.448960
- “CNN-based Approaches For Cross-Subject Classification in Motor Imagery: From the State-of-The-Art to DynamicNet” In 2021 IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology (CIBCB), 2021, pp. 1–7 DOI: 10.1109/CIBCB49929.2021.9562821
- François Chollet “Keras”, https://keras.io, 2015
- Matias Valdenegro “keras-uncertainty” In GitHub repository GitHub, https://github.com/mvaldenegro/keras-uncertainty, 2023
- “Deterministic neural networks with inductive biases capture epistemic and aleatoric uncertainty” In arXiv preprint arXiv:2102.11582 2, 2021
- “Understanding Measures of Uncertainty for Adversarial Example Detection”, 2018 arXiv:1803.08533 [stat.ML]
- “Evaluating and boosting uncertainty quantification in classification” In arXiv preprint arXiv:1909.06030, 2019
- “Deep deterministic uncertainty: A simple baseline” In arXiv preprint arXiv:2102.11582, 2021
- Matias Valdenegro-Toro and Daniel Saromo Mori “A deeper look into aleatoric and epistemic uncertainty disentanglement” In 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), 2022, pp. 1508–1516 IEEE
- “What uncertainties do we need in bayesian deep learning for computer vision?” In Advances in neural information processing systems 30, 2017
Sponsor
Paper Prompts
Sign up for free to create and run prompts on this paper using GPT-5.
Top Community Prompts
Collections
Sign up for free to add this paper to one or more collections.