Search

Spiking Neural Network (SNN) has been recognized as the third generation of neural networks. Conventionally, a SNN can be converted from a pre-trained Artificial Neural Network (ANN) with less computation and memory than training from scratch. But, these converted SNNs are vulnerable to adversarial attacks. Numerical experiments demonstrate that the SNN trained by optimizing the loss function will be more adversarial robust, but the theoretical analysis for the mechanism of robustness is lacking. In this paper, we provide a theoretical explanation by analyzing the expected risk function. Starting by modeling the stochastic process introduced by the Poisson encoder, we prove that there is a positive semidefinite regularizer. Perhaps surprisingly, this regularizer can make the gradients of the output with respect to input closer to zero, thus resulting in inherent robustness against adversarial attacks. Extensive experiments on the CIFAR10 and CIFAR100 datasets support our point of view. For example, we find that the sum of squares of the gradients of the converted SNNs is 13 ∼ 160 times that of the trained SNNs. And, the smaller the sum of the squares of the gradients, the smaller the degradation of accuracy under adversarial attack. • The adversarial robustness of SNNs based on rate encoder is explained mathematically. • A regularizer distinguishes the directly trained SNN from the converted SNN. • The coding length can also change the gradient of the input to affect the robustness. • We compare more types of the trained SNN with the converted SNN. [ABSTRACT FROM AUTHOR]

Visualization from trained deep neural networks has drawn massive public attention in recent. One of the visualization approaches is to train images maximizing the activation of specific neurons. However, directly maximizing the activation would lead to unrecognizable images, which cannot provide any meaningful information. In this paper, we introduce a simple but effective technique to constrain the optimization route of the visualization. By adding two totally inverse transformations, image blurring and deblurring, to the optimization procedure, recognizable images can be created. Our algorithm is good at extracting the details in the images, which are usually filtered by previous methods in the visualizations. Extensive experiments on AlexNet, VGGNet and GoogLeNet illustrate that we can better understand the neural networks utilizing the knowledge obtained by the visualization. [ABSTRACT FROM AUTHOR]

Channel Pruning is one of the most widespread techniques used to compress deep neural networks while maintaining their performances. Currently, a typical pruning algorithm leverages neural architecture search to directly find networks with a configurable width, the key step of which is to identify representative subnet for various pruning ratios by training a supernet. However, current methods mainly follow a serial training strategy to optimize supernet, which is very time-consuming. In this work, we introduce PSE-Net, a novel parallel-subnets estimator for efficient channel pruning. Specifically, we propose a parallel-subnets training algorithm that simulate the forward–backward pass of multiple subnets by droping extraneous features on batch dimension, thus various subnets could be trained in one round. Our proposed algorithm facilitates the efficiency of supernet training and equips the network with the ability to interpolate the accuracy of unsampled subnets, enabling PSE-Net to effectively evaluate and rank the subnets. Over the trained supernet, we develop a prior-distributed-based sampling algorithm to boost the performance of classical evolutionary search. Such algorithm utilizes the prior information of supernet training phase to assist in the search of optimal subnets while tackling the challenge of discovering samples that satisfy resource constraints due to the long-tail distribution of network configuration. Extensive experiments demonstrate PSE-Net outperforms previous state-of-the-art channel pruning methods on the ImageNet dataset while retaining superior supernet training efficiency. For example, under 300M FLOPs constraint, our pruned MobileNetV2 achieves 75.2% Top-1 accuracy on ImageNet dataset, exceeding the original MobileNetV2 by 2.6 units while only cost 30%/16% times than BCNet/AutoAlim. • We propose PSE-Net, a novel parallel-subnets estimator for efficient channel pruning. • The prior-distributed-based sampling algorithm to facilitate subnets search. • 30%/16% total time for estimator training compared with BCNet/AutoSlim. • Exceeds the SOTAs on ImageNet and downstream tasks for a variety of FLOPs budgets. [ABSTRACT FROM AUTHOR]