SPIDEN: deep Spiking Neural Networks for efficient image denoising.

In recent years, Deep Convolutional Neural Networks (DCNNs) have outreached the performance of classical algorithms for image restoration tasks. However, most of these methods are not suited for computational efficiency. In this work, we investigate Spiking Neural Networks (SNNs) for the specific and uncovered case of image denoising, with the goal of reaching the performance of conventional DCNN while reducing the computational cost. This task is challenging for two reasons. First, as denoising is a regression task, the network has to predict a continuous value (i.e., the noise amplitude) for each pixel of the image, with high precision. Moreover, state of the art results have been obtained with deep networks that are notably difficult to train in the spiking domain. To overcome these issues, we propose a formal analysis of the information conversion processing carried out by the Integrate and Fire (IF) spiking neurons and we formalize the trade-off between conversion error and activation sparsity in SNNs. We then propose, for the first time, an image denoising solution based on SNNs. The SNN networks are trained directly in the spike domain using surrogate gradient learning and backpropagation through time. Experimental results show that the proposed SNN provides a level of performance close to the state of the art with CNN based solutions. Specifically, our SNN achieves 30.18 dB of signal-to-noise ratio on the Set12 dataset, which is only 0.25 dB below the performance of the equivalent DCNN. Moreover we show that this performance can be achieved with low latency, i.e., using few timesteps, and with a significant level of sparsity. Finally, we analyze the energy consumption for different network latencies and network sizes. We show that the energy consumption of SNNs increases with longer latencies, making them more energy efficient compared to CNNs only for very small inference latencies. However, we also show that by increasing the network size, SNNs can provide competitive denoising performance while reducing the energy consumption by 20%.

Unsupervised Anomaly Detection for Cars CAN Sensors Time Series Using Small Recurrent and Convolutional Neural Networks.

Predictive maintenance in the car industry is an active field of research for machine learning and anomaly detection. The capability of cars to produce time series data from sensors is growing as the car industry is heading towards more connected and electric vehicles. Unsupervised anomaly detectors are therefore very adapted to process those complex multidimensional time series and highlight abnormal behaviors. We propose to use recurrent and convolutional neural networks based on unsupervised anomaly detectors with simple architectures on real, multidimensional time series generated by the car sensors and extracted from the Controller Area Network bus (CAN). Our method is then evaluated through known specific anomalies. As the computational costs of Machine Learning algorithms are a rising issue regarding embedded scenarios such as car anomaly detection, we also focus on creating anomaly detectors that are as small as possible. Using a state-of-the-art methodology incorporating a time series predictor and a prediction-error-based anomaly detector, we show that we can obtain roughly the same anomaly detection performance with smaller predictors, reducing parameters and calculations by up to 23% and 60%, respectively. Finally, we introduce a method to correlate variables with specific anomalies by using anomaly detector results and labels.

Trainable quantization for Speedy Spiking Neural Networks.

Spiking neural networks are considered as the third generation of Artificial Neural Networks. SNNs perform computation using neurons and synapses that communicate using binary and asynchronous signals known as spikes. They have attracted significant research interest over the last years since their computing paradigm allows theoretically sparse and low-power operations. This hypothetical gain, used from the beginning of the neuromorphic research, was however limited by three main factors: the absence of an efficient learning rule competing with the one of classical deep learning, the lack of mature learning framework, and an important data processing latency finally generating energy overhead. While the first two limitations have recently been addressed in the literature, the major problem of latency is not solved yet. Indeed, information is not exchanged instantaneously between spiking neurons but gradually builds up over time as spikes are generated and propagated through the network. This paper focuses on quantization error, one of the main consequence of the SNN discrete representation of information. We argue that the quantization error is the main source of accuracy drop between ANN and SNN. In this article we propose an in-depth characterization of SNN quantization noise. We then propose a end-to-end direct learning approach based on a new trainable spiking neural model. This model allows adapting the threshold of neurons during training and implements efficient quantization strategies. This novel approach better explains the global behavior of SNNs and minimizes the quantization noise during training. The resulting SNN can be trained over a limited amount of timesteps, reducing latency, while beating state of the art accuracy and preserving high sparsity on the main datasets considered in the neuromorphic community.

Quantization and Deployment of Deep Neural Networks on Microcontrollers.

Embedding Artificial Intelligence onto low-power devices is a challenging task that has been partly overcome with recent advances in machine learning and hardware design. Presently, deep neural networks can be deployed on embedded targets to perform different tasks such as speech recognition, object detection or Human Activity Recognition. However, there is still room for optimization of deep neural networks onto embedded devices. These optimizations mainly address power consumption, memory and real-time constraints, but also an easier deployment at the edge. Moreover, there is still a need for a better understanding of what can be achieved for different use cases. This work focuses on quantization and deployment of deep neural networks onto low-power 32-bit microcontrollers. The quantization methods, relevant in the context of an embedded execution onto a microcontroller, are first outlined. Then, a new framework for end-to-end deep neural networks training, quantization and deployment is presented. This framework, called MicroAI, is designed as an alternative to existing inference engines (TensorFlow Lite for Microcontrollers and STM32Cube.AI). Our framework can indeed be easily adjusted and/or extended for specific use cases. Execution using single precision 32-bit floating-point as well as fixed-point on 8- and 16 bits integers are supported. The proposed quantization method is evaluated with three different datasets (UCI-HAR, Spoken MNIST and GTSRB). Finally, a comparison study between MicroAI and both existing embedded inference engines is provided in terms of memory and power efficiency. On-device evaluation is done using ARM Cortex-M4F-based microcontrollers (Ambiq Apollo3 and STM32L452RE).