Hyperdimensional Computing for Blind and One-Shot Classification of EEG Error-Related Potentials

Rahimi, Abbas; Tchouprina, Artiom; Kanerva, Pentti; Millán, José del R.; Rabaey, Jan M.

doi:10.1007/s11036-017-0942-6

Cited by 51 publications

(41 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We also observe that HD computing quickly learns from one or two seizures and perfectly detects unseen seizures for the majority of patients (10 out of 16) [19]. Similar benefit is observed for the EEG ERP classification task [17], [18]: the HD algorithm learns ≈ 3× faster by using only 34% of training trials while maintaining an average accuracy of 70.5%, which is higher than the state-of-the-art classifier using the full set of training trials.…”

Section: Learning Is One-shot Fast and Computationally Balanced supporting

confidence: 71%

“…Besides, the learned HD vectors can be analyzed to identify what electrodes provide meaningful data for the classification. It has been shown that instead of asking for the domain expert knowledge, HD computing can identify the same subset of electrodes as relevant by measuring the relative distances between the learned prototype HD vectors [18]. Producing such transparent codes also enables verification of the learned model [52], and is in sharp contrast to blind application of conventional learning methods that produce a "black box.…”

Section: B Learning Transparent Codes With Interpretable Featuresmentioning

confidence: 99%

“…In this article, we first focus on the key properties of HD computing resulting from the application of a well-defined set of arithmetic operations on HD vectors. Key properties that are examined include: generality, scalability, analyzability, one-shot learning, energy efficiency, natural performance without domain expert knowledge and less preprocessing [5], [14], [15], [16], [17], [18], [19]. These leading properties take HD computing beyond the typical text and language applications [20], [21], [22], [23], [24], [25], [26], and make it a prime candidate for a new category of applications: integrated learning-based wearable/implantable devices, for which biosignal training data sets are small, individual variability is significant, privacy, latency, and energy efficiency demands are tight [27], [28], [29], [30].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Efficient Biosignal Processing Using Hyperdimensional Computing: Network Templates for Combined Learning and Classification of ExG Signals

et al. 2019

Self Cite

View full text Add to dashboard Cite

Recognizing the very size of the brain's circuits, hyperdimensional (HD) computing can model neural activity patterns with points in a HD space, that is, with HD vectors. Key examined properties of HD computing include: a versatile set of arithmetic operations on HD vectors, generality, scalability, analyzability, one-shot learning, and energy efficiency. These make it a prime candidate for efficient biosignal processing where signals are noisy and nonstationary, training data sets are not huge, individual variability is significant, and energy efficiency constraints are tight. Purely based on native HD computing operators, we describe a combined method for multiclass learning and classification of various ExG biosignals such as electromyography (EMG), electroencephalography (EEG), and electrocorticography (ECoG). We develop a full set of HD network templates that comprehensively encode body potentials and brain neural activity recorded from different electrodes into a single HD vector without requiring domain expert knowledge or ad-hoc electrode selection process. Such encoded HD vector is processed as a single unit for fast one-shot learning, and robust classification. It can be interpreted to identify the most useful features as well. Compared to state-of-the-art counterparts, HD computing enables online, incremental, and fast learning as it demands less than a third as much training data as well as less preprocessing.

show abstract

Section: Learning Is One-shot Fast and Computationally Balanced supporting

confidence: 71%

Section: B Learning Transparent Codes With Interpretable Featuresmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Efficient Biosignal Processing Using Hyperdimensional Computing: Network Templates for Combined Learning and Classification of ExG Signals

et al. 2019

Self Cite

View full text Add to dashboard Cite

show abstract

“…Based on the use of these MAP operations, an encoder can be designed for various tasks, e.g., EMG [20], [22], [49], EEG [23], [50], ECoG [51], ExG [45], or in general pattern processing [52]. The encoder emits a hypervector representing the event of interest that is then fed into an associative memory (AM) for training and inference.…”

Section: B Hyperdimensional Computingmentioning

confidence: 99%

“…The proposed system is highly flexible and versatile, being capable of adapting to the changes in the initial setup and other interferences that can deteriorate the performance. Moreover, it can be easily adapted to other kinds of applications, where a higher number of acquisition channels is required, both for EEG [23] and EMG [22] based applications. The system reaches 90.40% accuracy on 11 gestures recognition, comparable to the SoA systems, within an energy budget of just 10.04mJ for the onchip training and 83.20µJ for the recognition (inference), leading to an average power consumption of 10.04mW reaching 29 hours battery life with a 100mAh battery.…”

Section: Introductionmentioning

confidence: 99%

Online Learning and Classification of EMG-Based Gestures on a Parallel Ultra-Low Power Platform Using Hyperdimensional Computing

Benatti

Montagna

Kartsch

et al. 2019

IEEE Trans. Biomed. Circuits Syst.

Self Cite

View full text Add to dashboard Cite

This work presents a wearable EMG gesture recognition system based on the hyperdimensional (HD) computing paradigm, running on a programmable Parallel Ultra-Low-Power (PULP) platform. The processing chain includes efficient on-chip training, which leads to a fully embedded implementation with no need to perform any offline training on a personal computer. The proposed solution has been tested on 10 subjects in a typical gesture recognition scenario achieving 85% average accuracy on 11 gestures recognition, which is aligned with the State-Of-the-Art (SoA), with the unique capability of performing online learning. Furthermore, by virtue of the Hardware (HW) friendly algorithm and of the efficient PULP System-on-Chip (SoC) (Mr. Wolf) used for prototyping and evaluation, the energy budget required to run the learning part with 11 gestures is 10.04mJ, and 83.2µJ per classification. The system works with a average power consumption of 10.4mW in classification, ensuring around 29h of autonomy with a 100mAh battery. Finally, the scalability of the system is explored by increasing the number of channels (up-to 256 electrodes), demonstrating the suitability of our approach as universal, energy-efficient biopotential wearable recognition framework.

show abstract

High Dimensional Computing Approach to Detection and Learning Gesture Biometrics

Liu,

Casey,

Melaragno

2024

Lecture Notes in Networks and Systems

View full text Add to dashboard Cite

Hyperdimensional Computing for Blind and One-Shot Classification of EEG Error-Related Potentials

Cited by 51 publications

References 14 publications

Efficient Biosignal Processing Using Hyperdimensional Computing: Network Templates for Combined Learning and Classification of ExG Signals

Efficient Biosignal Processing Using Hyperdimensional Computing: Network Templates for Combined Learning and Classification of ExG Signals

Online Learning and Classification of EMG-Based Gestures on a Parallel Ultra-Low Power Platform Using Hyperdimensional Computing

High Dimensional Computing Approach to Detection and Learning Gesture Biometrics

Contact Info

Product

Resources

About