Search

Hyperdimensional computing (HDC) is a brain-inspired computing paradigm that operates on pseudo-random hypervectors to perform high-accuracy classifications for biomedical applications. The energy efficiency of prior HDC processors for this computationally minimal algorithm is dominated by costly hypervector memory storage, which grows linearly with the number of sensors. To address this, the memory is replaced with a light-weight cellular automaton for on-the-fly hypervector generation. The use of this technique is explored in conjunction with vector folding for various real-time classification latencies in post-layout simulation on an emotion recognition dataset with 200 channels. The proposed architecture achieves 39.1 nJ/prediction; a 4.9× energy efficiency improvement, 9.5× per channel, over the state-of-the-art HDC processor. At maximum throughput, the architecture achieves a 10.7× improvement, 33.5× per channel. An optimized support vector machine (SVM) processor is designed in this work for the same use-case. HDC is 9.5× more energy-efficient than the SVM, paving the way for it to become the paradigm of choice for high-accuracy, on-board biosignal classification.

Closed-loop neuromodulation systems aim to treat a variety of neurological conditions by delivering and adjusting therapeutic electrical stimulation in response to a patient's neural state, recorded in real time. Existing systems are limited by low channel counts, lack of algorithmic flexibility, and the distortion of recorded signals by large and persistent stimulation artefacts. Here, we describe an artefact-free wireless neuromodulation device that enables research applications requiring high-throughput data streaming, low-latency biosignal processing, and simultaneous sensing and stimulation. The device is a miniaturized neural interface capable of closed-loop recording and stimulation on 128 channels, with on-board processing to fully cancel stimulation artefacts. In addition, it can detect neural biomarkers and automatically adjust stimulation parameters in closed-loop mode. In a behaving non-human primate, the device enabled long-term recordings of local field potentials and the real-time cancellation of stimulation artefacts, as well as closed-loop stimulation to disrupt movement preparatory activity during a delayed-reach task. The neuromodulation device may help advance neuroscientific discovery and preclinical investigations of stimulation-based therapeutic interventions.

The mathematical properties of high-dimensional (HD) spaces show remarkable agreement with behaviors controlled by the brain. Computing with HD vectors, referred to as "hypervectors," is a brain-inspired alternative to computing with numbers. HD computing is characterized by generality, scalability, robustness, and fast learning, making it a prime candidate for utilization in application domains such as brain–computer interfaces. We describe the use of HD computing to classify electroencephalography (EEG) error-related potentials for noninvasive brain–computer interfaces. Our algorithm naturally encodes neural activity recorded from 64 EEG electrodes to a single temporal–spatial hypervector without requiring any electrode selection process. This hypervector represents the event of interest, can be analyzed to identify the most discriminative electrodes, and is used for recognition of the subject's intentions. Using the full set of training trials, HD computing achieves on average 5% higher single-trial classification accuracy compared to a conventional machine learning method on this task (74.5% vs. 69.5%) and offers further advantages: (1) Our algorithm learns fast: using only 34% of training trials it achieves an average accuracy of 70.5%, surpassing the conventional method. (2) Conventional method requires prior domain expert knowledge, or a separate process, to carefully select a subset of electrodes for a subsequent preprocessor and classifier, whereas our algorithm blindly uses all 64 electrodes, tolerates noises in data, and the resulting hypervector is intrinsically clustered into HD space; in addition, most preprocessing of the electrode signal can be eliminated while maintaining an average accuracy of 71.7%. [ABSTRACT FROM AUTHOR]

Presents a study which established a relationship between the basic properties of linear computations and optimizing transformations. Use of transformations in computer-aided design; Three algorithms discussed in the study.