Multitask Learning for Brain-Computer Interfaces

By Morteza Alamgir, Moritz Grosse-Wentrup, Yasemin Altun. 2010

98 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Neuroscience

Citations in the corpus

An Overview of Multi-task Learning (2018, background): “Zhang and Yang 39 populations [104], construction of personalized brain–computer interfaces [105], MHC-I binding prediction [106], splice-site prediction [106], protein subcellular location prediction [107], Alzheimer’s disease assessment scale cognitive subscale [108], prediction of cognitive…”
A Survey on Multi-Task Learning (2017, methodology): “In [185], a multi-task model by sharing a Gaussian prior on model parameters of different tasks is to construct personalized brain-computer interfaces.”
Online and Offline Domain Adaptation for Reducing BCI Calibration Effort (2017, methodology): “For each headset, we first bandpassed the EEG signals to [1, 50] Hz, then downsampled them to 64 Hz, performed average reference, and next epoched them to the [0, 0.”
Signal Processing Approaches to Minimize or Suppress Calibration Time in Oscillatory Activity-Based Brain–Computer Interfaces (2015, background): “This has been explored successfully for multi-task linear classifier learning [38] and multi-task CSP learning [39]–[41]. c) Domain adaptation: Domain adaptation consists of transferring classifiers or features (and in particular spatial filters), which were optimized on other users, to the target…”
Subject-Independent Brain–Computer Interfaces Based on Deep Convolutional Neural Networks (2019, background): “In addition, studies that utilize minimum calibration data have been conducted [34]–[39].”
Feature-based transfer learning with real-world applications (2010, methodology): “[3] applied multi-task learning techniques to solve braincomputer interfaces problems.”
Multisubject Learning for Common Spatial Patterns in Motor-Imagery BCI (2011, background): “In [12], the authors employed this concept to learn a classiﬁer that was able to learn from multiple subjects, leading to an algorithm that performed well on new subjects even without training.”
Transfer Learning in Brain-Computer Interfaces (2015, methodology): “ble in the age of big data. It is likely that all the methods presented here would perform better if prior knowledge had been incorporated into choosing the feature space. For example, Alamgir et al. [33] use data only from the electrodes directly over the motor cortex. Indeed, given a small feature”
Regularizing Common Spatial Patterns to Improve BCI Designs: Unified Theory and New Algorithms (2011, background): “of spatial filters learnt across multiple subjects would be used as prior information, in a similar flavor as what has been done in [31] for learning linear classifiers.”