Live Demonstration: Face Recognition on an Ultra-Low Power Event-Driven Convolutional Neural Network ASIC

Liu, Qian; Richter, Ole; Nielsen, Charlotte; Sheik, Sadique; Indiveri, Giacomo; Qiao, Ning

doi:10.1109/cvprw.2019.00213

Cited by 18 publications

(16 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The ultra-low power consumption of mixed-signal neuromorphic chips make them suitable for edge-applications, such as always-on voice detection [32], vibration monitoring [33] or always- on face recognition [34]. For this reason, we consider two compact network architectures in our experiments: A Spiking Recurrent Neural Network (SRNN) with roughly 65k trainable parameters and a conventional CNN with roughly 500k trainable parameters (see S1 for more information).…”

Section: Resultsmentioning

confidence: 99%

Network insensitivity to parameter noise via adversarial regularization

Büchel¹,

Fynn²,

Muir³

2021

Preprint

View full text Add to dashboard Cite

Neuromorphic neural network processors, in the form of compute-in-memory crossbar arrays of memristors, or in the form of subthreshold analog and mixed-signal ASICs, promise enormous advantages in compute density and energy efficiency for NN-based ML tasks. However, these technologies are prone to computational non-idealities, due to process variation and intrinsic device physics. This degrades the task performance of networks deployed to the processor, by introducing parameter noise into the deployed model. While it is possible to calibrate each device, or train networks individually for each processor, these approaches are expensive and impractical for commercial deployment. Alternative methods are therefore needed to train networks that are inherently robust against parameter variation, as a consequence of network architecture and parameters. We present a new adversarial network optimisation algorithm that attacks network parameters during training, and promotes robust performance during inference in the face of parameter variation. Our approach introduces a regularization term penalising the susceptibility of a network to weight perturbation. We compare against previous approaches for producing parameter insensitivity such as dropout, weight smoothing and introducing parameter noise during training. We show that our approach produces models that are more robust to targeted parameter variation, and equally robust to random parameter variation. Our approach finds minima in flatter locations in the weight-loss landscape compared with other approaches, highlighting that the networks found by our technique are less sensitive to parameter perturbation. Our work provides an approach to deploy neural network architectures to inference devices that suffer from computational non-idealities, with minimal loss of performance. This method will enable deployment at scale to novel energy-efficient computational substrates, promoting cheaper and more prevalent edge inference.Preprint. Under review.

show abstract

Section: Resultsmentioning

confidence: 99%

Network insensitivity to parameter noise via adversarial regularization

Büchel¹,

Fynn²,

Muir³

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…After training, we tested our trained weights on spiking network simulations. Unlike tests done on analog networks, these are time-dependent simulations, which fully account for the time dynamics of the input spike trains, and closely mimic the behavior of a neuromorphic hardware implementation, like DynapCNN (Liu et al, 2019 ). Our simulations are written using the Sinabs Python library 2 , which uses non-leaky integrate-and-fire neurons with a linear response function.…”

Section: Methodsmentioning

confidence: 99%

Optimizing the Energy Consumption of Spiking Neural Networks for Neuromorphic Applications

Sorbaro

Liu²,

Bortone³

et al. 2020

Front. Neurosci.

Self Cite

View full text Add to dashboard Cite

In the last few years, spiking neural networks (SNNs) have been demonstrated to perform on par with regular convolutional neural networks. Several works have proposed methods to convert a pre-trained CNN to a Spiking CNN without a significant sacrifice of performance. We demonstrate first that quantization-aware training of CNNs leads to better accuracy in SNNs. One of the benefits of converting CNNs to spiking CNNs is to leverage the sparse computation of SNNs and consequently perform equivalent computation at a lower energy consumption. Here we propose an optimization strategy to train efficient spiking networks with lower energy consumption, while maintaining similar accuracy levels. We demonstrate results on the MNIST-DVS and CIFAR-10 datasets.

show abstract

“…Spiking neuromorphic implementations include large-scale simulation of neuronal networks for neuroscience research (Furber et al, 2012) and lowpower real-world deployments of machine learning algorithms. In particular, convolutional neural network (CNN) architectures, used for computer vision, have been run on neuromorphic chips such as IBM 's TrueNorth (Esser et al, 2016), Intel's Loihi (Davies et al, 2018) and SynSense's Speck and Dynap-CNN hardware (Liu et al, 2019a). The full pipeline of event-based sensors that output sparse data, stateful spiking neural networks which extract semantic meaning and asynchronous hardware backends allows for large gains in power-efficiency when compared to conventional systems.…”

Section: What Is Event-based Sensing?mentioning

confidence: 99%

Adversarial Attacks on Spiking Convolutional Neural Networks for Event-based Vision

Büchel¹,

Lenz²,

Hu³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Event-based sensing using dynamic vision sensors is gaining traction in lowpower vision applications. Spiking neural networks work well with the sparse nature of event-based data and suit deployment on low-power neuromorphic hardware. Being a nascent field, the sensitivity of spiking neural networks to potentially malicious adversarial attacks has received very little attention so far. In this work, we show how white-box adversarial attack algorithms can be adapted to the discrete and sparse nature of event-based visual data, and to the continuous-time setting of spiking neural networks. We test our methods on the N-MNIST and IBM Gestures neuromorphic vision datasets and show adversarial perturbations achieve a high success rate, by injecting a relatively small number of appropriately placed events. We also verify, for the first time, the effectiveness of these perturbations directly on neuromorphic hardware. Finally, we discuss the properties of the resulting perturbations and possible future directions.

show abstract

Live Demonstration: Face Recognition on an Ultra-Low Power Event-Driven Convolutional Neural Network ASIC

Cited by 18 publications

References 4 publications

Network insensitivity to parameter noise via adversarial regularization

Network insensitivity to parameter noise via adversarial regularization

Optimizing the Energy Consumption of Spiking Neural Networks for Neuromorphic Applications

Adversarial Attacks on Spiking Convolutional Neural Networks for Event-based Vision

Contact Info

Product

Resources

About