Comparison of Artificial and Spiking Neural Networks on Digital Hardware

Davidson, Simon; Furber, Steve

doi:10.3389/fnins.2021.651141

Cited by 83 publications

(39 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, previous work relied on rate-based or time-to-first-spike coding schemes. Here, we expanded ITL techniques into the realm of surrogate gradient learning, which flexibly interpolates between rate- and timing-based coding schemes on multilayer and recurrent architectures, thereby simultaneously improving performance and energy efficiency, while also being conducive for fast inference ( 24 ).…”

Section: Discussionmentioning

confidence: 99%

“…First, one has to overcome the binary nature of spikes, which impedes vanilla gradient descent ( 21 – 23 ). Second, training has to ensure sparse spiking activity to exploit the superior power efficiency of SNN processing ( 24 , 25 ). Finally, training has to achieve all of the above while coping with analog hardware imperfections inevitably tied to their manufacturing process.…”

mentioning

confidence: 99%

See 1 more Smart Citation

Surrogate gradients for analog neuromorphic computing

Cramer

Billaudelle

Kanya

et al. 2022

Proc. Natl. Acad. Sci. U.S.A.

View full text Add to dashboard Cite

To rapidly process temporal information at a low metabolic cost, biological neurons integrate inputs as an analog sum, but communicate with spikes, binary events in time. Analog neuromorphic hardware uses the same principles to emulate spiking neural networks with exceptional energy efficiency. However, instantiating high-performing spiking networks on such hardware remains a significant challenge due to device mismatch and the lack of efficient training algorithms. Surrogate gradient learning has emerged as a promising training strategy for spiking networks, but its applicability for analog neuromorphic systems has not been demonstrated. Here, we demonstrate surrogate gradient learning on the BrainScaleS-2 analog neuromorphic system using an in-the-loop approach. We show that learning self-corrects for device mismatch, resulting in competitive spiking network performance on both vision and speech benchmarks. Our networks display sparse spiking activity with, on average, less than one spike per hidden neuron and input, perform inference at rates of up to 85,000 frames per second, and consume less than 200 mW. In summary, our work sets several benchmarks for low-energy spiking network processing on analog neuromorphic hardware and paves the way for future on-chip learning algorithms.

show abstract

Section: Discussionmentioning

confidence: 99%

mentioning

confidence: 99%

Surrogate gradients for analog neuromorphic computing

Cramer

Billaudelle

Kanya

et al. 2022

Proc. Natl. Acad. Sci. U.S.A.

View full text Add to dashboard Cite

show abstract

“…This is a simplification, because in reality the actual energy consumption also depends on the specific hardware that the networks run on. However, this is a widely adopted metric because it is simple to calculate and allows for an approximate comparison [49,14,50,33,51]. In [52], an overview over different neuromorphic architectures is given that shows a range of 2.8-360 pJ per synaptic operation, i.e., a factor of 100 between different hardware implementations is possible for SNNs.…”

Section: Appendixmentioning

confidence: 99%

Hybrid SNN-ANN: Energy-Efficient Classification and Object Detection for Event-Based Vision

Kugele,

Pfeil,

Pfeiffer

et al. 2021

Preprint

View full text Add to dashboard Cite

Event-based vision sensors encode local pixel-wise brightness changes in streams of events rather than full image frames and yield sparse, energy-efficient encodings of scenes, in addition to low latency, high dynamic range, and lack of motion blur. Recent progress in object recognition from event-based sensors has come from conversions of successful deep neural network architectures, which are trained with backpropagation. However, using these approaches for event streams requires a transformation to a synchronous paradigm, which not only loses computational efficiency, but also misses opportunities to extract spatio-temporal features. In this article we propose a hybrid architecture for end-to-end training of deep neural networks for event-based pattern recognition and object detection, combining a spiking neural network (SNN) backbone for efficient event-based feature extraction, and a subsequent classical analog neural network (ANN) head to solve synchronous classification and detection tasks. This is achieved by combining standard backpropagation with surrogate gradient training to propagate gradients inside the SNN layers. Hybrid SNN-ANNs can be trained without additional conversion steps, and result in highly accurate networks that are substantially more computationally efficient than their ANN counterparts. We demonstrate results on event-based classification and object detection datasets, in which only the architecture of the ANN heads need to be adapted to the tasks, and no conversion of the event-based input is necessary. Since ANNs and SNNs require different hardware paradigms to maximize their efficiency, we envision that SNN backbone and ANN head can be executed on different processing units, and thus analyze the necessary bandwidth to communicate between the two parts. Hybrid networks are promising architectures to further advance machine learning approaches for event-based vision, without having to compromise on efficiency.

show abstract

“…System efficiency comes from sensing and data processing. Unlike classical vision systems, neuromorphic systems try to efficiently capture a notion of seeing motion [5][6][7][8][9]. Bio-inspired learning methods, i.e., spiking neural networks (SNNs), address issues related to energy efficiency [5,[10][11][12][13][14][15][16][17][18][19][20][21][22].…”

Section: Introductionmentioning

confidence: 99%

Spiking Hyperdimensional Network: Neuromorphic Models Integrated with Memory-Inspired Framework

Zou,

Alimohamadi,

Imani

et al. 2021

Preprint

View full text Add to dashboard Cite

Recently, brain-inspired computing models have shown great potential to outperform today's deep learning solutions in terms of robustness and energy efficiency. Particularly, Spiking Neural Networks (SNNs) and HyperDimensional Computing (HDC) have shown promising results in enabling efficient and robust cognitive learning. Despite the success, these two brain-inspired models have different strengths. While SNN mimics the physical properties of the human brain, HDC models the brain on a more abstract and functional level. Their design philosophies demonstrate complementary patterns that motivate their combination. With the help of the classical psychological model on memory, we propose SpikeHD, the first framework that fundamentally combines Spiking neural network and hyperdimensional computing. SpikeHD generates a scalable and strong cognitive learning system that better mimics brain functionality. SpikeHD exploits spiking neural networks to extract low-level features by preserving the spatial and temporal correlation of raw event-based spike data. Then, it utilizes HDC to operate over SNN output by mapping the signal into high-dimensional space, learning the abstract information, and classifying the data. Our extensive evaluation on a set of benchmark classification problems shows that SpikeHD provides the following benefit compared to SNN architecture:(1) significantly enhance learning capability by exploiting two-stage information processing, (2) enables substantial robustness to noise and failure, and (3) reduces the network size and required parameters to learn complex information.

show abstract

Comparison of Artificial and Spiking Neural Networks on Digital Hardware

Cited by 83 publications

References 24 publications

Surrogate gradients for analog neuromorphic computing

Surrogate gradients for analog neuromorphic computing

Hybrid SNN-ANN: Energy-Efficient Classification and Object Detection for Event-Based Vision

Spiking Hyperdimensional Network: Neuromorphic Models Integrated with Memory-Inspired Framework

Contact Info

Product

Resources

About