Noise Injection Research Articles

Attack Deterministic Conditional Image Generative Models for Diverse and Controllable Generation

Existing generative adversarial network (GAN) based conditional image generative models typically produce fixed output for the same conditional input, which is unreasonable for highly subjective tasks, such as large-mask image inpainting or style transfer. On the other hand, GAN-based diverse image generative methods require retraining/fine-tuning the network or designing complex noise injection functions, which is computationally expensive, task-specific, or struggle to generate high-quality results. Given that many deterministic conditional image generative models have been able to produce high-quality yet fixed results, we raise an intriguing question: is it possible for pre-trained deterministic conditional image generative models to generate diverse results without changing network structures or parameters? To answer this question, we re-examine the conditional image generation tasks from the perspective of adversarial attack and propose a simple and efficient plug-in projected gradient descent (PGD) like method for diverse and controllable image generation. The key idea is attacking the pre-trained deterministic generative models by adding a micro perturbation to the input condition. In this way, diverse results can be generated without any adjustment of network structures or fine-tuning of the pre-trained models. In addition, we can also control the diverse results to be generated by specifying the attack direction according to a reference text or image. Our work opens the door to applying adversarial attack to low-level vision tasks, and experiments on various conditional image generation tasks demonstrate the effectiveness and superiority of the proposed method.

Investigation into vibration analysis for detecting faults in vehicle steering outer tie-rod

This study presents a novel fault detection method in car gear steering systems, employing MSC Adams and MATLAB simulations to analyze angular acceleration from the outer tie rod. The approach closely mimics real accelerometer data to differentiate between normal and faulty conditions, including wear and obstacle navigation. Emphasis is on noise robustness, utilizing advanced noise injection and denoising techniques. The efficacy of wavelet scattering, discrete wavelet transform (DWT) methods, and classifiers like Support Vector Machines (SVM) and Neural Networks (NN) is extensively evaluated. Among fifteen fault detection methods, the combination of wavelet scattering with Long Short-Term Memory (LSTM) Neural Networks, optimized with Adam tuning, is notably stable across four scenarios. The research highlights the importance of precise feature selection, employing techniques like Principal Component Analysis (PCA), Linear Discriminant Analysis (LDA), and Recursive Feature Elimination (RFE). This research significantly advances the reliability of autonomous driving systems and provides essential insights into fault detection in gear steering systems.

Joint Optimization of Beamforming and Noise Injection for Covert Downlink Transmissions in Cell-Free Internet of Things Networks

Joint Optimization of Beamforming and Noise Injection for Covert Downlink Transmissions in Cell-Free Internet of Things Networks

Improving throughput in Orbitrap using the ion funnel with orthogonal injection and observation of chemical noise in high-resolution MS instruments

Overall, the nanoflow chromatographic separation, nano-electrospray source, and Orbitrap Elite mass spectrometer with the dual-stage ion funnel that replaced the standard heated-cone inlet were used in the MS analysis of low-abundance peptide samples. The analytical advantage of the ion funnel was demonstrated in the LC-MS analysis of the HeLa protein digest mixture. Chemical noise patterns in Orbitrap, linear ion traps, and quadrupole time-of-flight mass spectrometers were compared. The chemical noise patterns in Orbitrap were found to be generally similar to those observed in linear ion traps and time-of-flight instruments, yet it was observed only under the condition of selecting a few Da mass range for the mass analysis. The Orbitrap's ultrahigh mass resolution enabled the intricate internal structure of the chemical noise to be uncovered. In addition to singly charged ions representing a bulk of chemical noise, doubly charged ions were also observed, though the intensity of the latter was found to be significantly lower, suggesting that chemical noise-limited detection levels in the MS analysis of low-abundance peptides could be significantly lower for 2+ or 3+ peptide ions compared to singly charged ones.

Privacy-Preserving Process Mining: A Blockchain-Based Privacy-Aware Reversible Shared Image Approach

ABSTRACT Deeper integration of cross-organizational business process sharing and process mining has advanced the Industrial Internet. Privacy breaches and data security risks limit its use. Scrambling or anonymizing event data frequently preserves privacy in established studies. The scrambling mechanism or random noise injection corrupts event log process information and lowers process mining outcomes. This research presents a blockchain-based privacy-aware reversible shared image approach using chaotic image and privacy-aware theory for privacy-preserving process mining. Avoiding data loss, disclosure concerns, correlation attacks, and encrypted sharing is possible with the method. First, process data is turned into color images with chaotic image encryption to safeguard privacy and allow reversible reproduction. Second, the on-chain-off-chain paradigm helps handle information lightly; finally, attribute encryption of multi-view event data for correlation resistance and on-demand data encryption sharing. Simulations on common datasets reveal that: 1. The system performance of the proposed method outperforms the baseline method by 57%. 2. The strategy greatly enhances categorical and numerical data privacy. 3. It performs better in event data privacy protection and process mining fitness and precision. The proposed method ensures the secure flow of cross-organizational information in the Industrial Internet and provides a novel privacy-secure computational approach for the growing Artificial Intelligence.

Gradient-descent hardware-aware training and deployment for mixed-signal neuromorphic processors

Mixed-signal neuromorphic processors provide extremely low-power operation for edge inference workloads, taking advantage of sparse asynchronous computation within spiking neural networks (SNNs). However, deploying robust applications to these devices is complicated by limited controllability over analog hardware parameters, as well as unintended parameter and dynamical variations of analog circuits due to fabrication non-idealities. Here we demonstrate a novel methodology for offline training and deployment of SNNs to the mixed-signal neuromorphic processor DYNAP-SE2. Our methodology applies gradient-based training to a differentiable simulation of the mixed-signal device, coupled with an unsupervised weight quantization method to optimize the network’s parameters. Parameter noise injection during training provides robustness to the effects of quantization and device mismatch, making the method a promising candidate for real-world applications under hardware constraints and non-idealities. This work extends Rockpool, an open-source deep-learning library for SNNs, with support for accurate simulation of mixed-signal SNN dynamics. Our approach simplifies the development and deployment process for the neuromorphic community, making mixed-signal neuromorphic processors more accessible to researchers and developers.

Data augmentation for invasive brain–computer interfaces based on stereo-electroencephalography (SEEG)

Objective. Deep learning is increasingly used for brain–computer interfaces (BCIs). However, the quantity of available data is sparse, especially for invasive BCIs. Data augmentation (DA) methods, such as generative models, can help to address this sparseness. However, all the existing studies on brain signals were based on convolutional neural networks and ignored the temporal dependence. This paper attempted to enhance generative models by capturing the temporal relationship from a time-series perspective. Approach. A conditional generative network (conditional transformer-based generative adversarial network (cTGAN)) based on the transformer model was proposed. The proposed method was tested using a stereo-electroencephalography (SEEG) dataset which was recorded from eight epileptic patients performing five different movements. Three other commonly used DA methods were also implemented: noise injection (NI), variational autoencoder (VAE), and conditional Wasserstein generative adversarial network with gradient penalty (cWGANGP). Using the proposed method, the artificial SEEG data was generated, and several metrics were used to compare the data quality, including visual inspection, cosine similarity (CS), Jensen–Shannon distance (JSD), and the effect on the performance of a deep learning-based classifier. Main results. Both the proposed cTGAN and the cWGANGP methods were able to generate realistic data, while NI and VAE outputted inferior samples when visualized as raw sequences and in a lower dimensional space. The cTGAN generated the best samples in terms of CS and JSD and outperformed cWGANGP significantly in enhancing the performance of a deep learning-based classifier (each of them yielding a significant improvement of 6% and 3.4%, respectively). Significance. This is the first time that DA methods have been applied to invasive BCIs based on SEEG. In addition, this study demonstrated the advantages of the model that preserves the temporal dependence from a time-series perspective.

Lab to Multiscene Generalization for Non-Line-of-Sight Identification With Small-Scale Datasets

Ultra-wideband (UWB) wireless indoor positioning systems rely on time of flight (TOF) to estimate distances but can be biased and miscalculated due to non-line-of-sight (NLOS) transmission channels in complex environments. Therefore, to remove errors, several machine learning techniques have been proposed for identifying NLOS signals from Channel Impulse Responses (CIRs). However, as CIR signals could be heavily influenced by various environments, current NLOS classifiers are not universal to provide satisfactory accuracy for new scenarios and require detailed measurements on a large number of CIRs for training. Hence, we propose a generalization method based on data augmentation via noise injection and transfer learning to allow the deep neural network (DNN) trained under a lab condition to be applied to various and even harsh practical scenarios with the need to measure massive training data minimized. This paper presents the first demonstration that it is effective to utilize a lab-based pre-trained DNN for real-world transfer and white Gaussian noise data augmentation for ML-based NLOS identification on UWB CIRs to address the problem when it is not feasible to measure sufficient training data. Our testing results show that in two scenarios, corridor and parking lot, with only 50 CIR signals as the training set, the accuracy of the NLOS identification model after applying the proposed method is increased from 84.4% to 98.8% and from 81.1% to 97.1%, respectively.

Numerical stability of DeepGOPlus inference.

Convolutional neural networks (CNNs) are currently among the most widely-used deep neural network (DNN) architectures available and achieve state-of-the-art performance for many problems. Originally applied to computer vision tasks, CNNs work well with any data with a spatial relationship, besides images, and have been applied to different fields. However, recent works have highlighted numerical stability challenges in DNNs, which also relates to their known sensitivity to noise injection. These challenges can jeopardise their performance and reliability. This paper investigates DeepGOPlus, a CNN that predicts protein function. DeepGOPlus has achieved state-of-the-art performance and can successfully take advantage and annotate the abounding protein sequences emerging in proteomics. We determine the numerical stability of the model's inference stage by quantifying the numerical uncertainty resulting from perturbations of the underlying floating-point data. In addition, we explore the opportunity to use reduced-precision floating point formats for DeepGOPlus inference, to reduce memory consumption and latency. This is achieved by instrumenting DeepGOPlus' execution using Monte Carlo Arithmetic, a technique that experimentally quantifies floating point operation errors and VPREC, a tool that emulates results with customizable floating point precision formats. Focus is placed on the inference stage as it is the primary deliverable of the DeepGOPlus model, widely applicable across different environments. All in all, our results show that although the DeepGOPlus CNN is very stable numerically, it can only be selectively implemented with lower-precision floating-point formats. We conclude that predictions obtained from the pre-trained DeepGOPlus model are very reliable numerically, and use existing floating-point formats efficiently.

Community detection in decentralized social networks with local differential privacy

Community detection under local differential privacy has been a research hotspot recently. Most existing methods primarily rely on the edge local differential privacy model (edge-LDP), compromising privacy strength for high utility. In addition, they adopt a way of agglomerating nodes one by one to construct communities (or clusters) through multiple iterations, suffering from significant noise and low detection efficiency. Moreover, the error accumulation caused by large-scale iterations is not conducive to detection accuracy. To address these issues, we propose a privacy-preserving community detection method, CD-LDP, based on node-LDP, achieving efficient and accurate community construction by aggregating grouped units instead of individual nodes and delimiting the interaction to three rounds. Specifically, we propose a privacy-preserving global social network construction method to gain connectivity information among all nodes through two rounds of interactions. In addition, a novel perturbation mechanism based on the star social network structure is devised, which improves the accuracy of the global social network by reducing the probability of generating fake edges. Secondly, we design a privacy-preserving node grouping approach leveraging clustering coefficients and Haar discrete wavelet transform in the third round, achieving low noise injection and high-quality node grouping through lossless dimensionality reduction. Lastly, we develop a fast unit merging algorithm combining modularity and Huffman trees to realize efficient and accurate community building by reducing the aggregation possibility of nodes with low similarity. Theoretical analysis and experimentation on real-world datasets testify that our solution can efficiently extract high-quality community structures while satisfying node-LDP.

Semi-supervised possibilistic c-means clustering algorithm based on feature weights for imbalanced data

The possibilistic c-means clustering (PCM) algorithm improves the robustness of fuzzy c-means clustering (FCM) to noise and outliers by releasing the probabilistic constraint of memberships. The semi-supervised possibilistic c-means clustering (SSPCM) algorithm improves the clustering effect on datasets with imbalanced sizes by introducing a small amount of label information. However, the traditional semi-supervised algorithm still faces the problem of low utilization of supervision information for datasets with large differences in sample sizes. Moreover, the Euclidean distance, which treats features equally, cannot handle feature-imbalanced data. Therefore, this paper proposes a semi-supervised possibilistic c-means clustering algorithm based on feature weights (FW-SSPCM) by introducing the ideas of supervised centers. First, the algorithm introduces the supervised center into the objective function of the SSPCM to improve the utilization rate of supervision information and thus guide the center iteration of small clusters. Second, the feature weighting strategy is introduced in the objective function to adaptively assign feature weights according to the importance of different features in different clusters, thus improving the adaptability of the algorithm to feature-imbalanced datasets. In addition, to improve the robustness of the antinoise effect and retain additional image details, a new image segmentation algorithm based on FW-SSPCM and local information (LFW-SSPCM) is proposed by introducing local spatial information obtained by bilateral filtering. Finally, through clustering experiments on synthetic data, UCI datasets and on color images characteristic of multiple features, including imbalanced sizes, imbalanced features and strong noise injection, the clustering performances of the proposed FW-SSPCM and LFW-SSPCM proposed in this paper are significantly better than those of several related clustering algorithms.

Chaotic neural network quantization and its robustness against adversarial attacks

Achieving robustness against adversarial attacks while maintaining high accuracy remains a critical challenge in neural networks. Parameter quantization is one of the main approaches used to compress deep neural networks to have less inference time and less storage memory size. However, quantization causes severe degradation in accuracy and consequently in model robustness. This work investigates the efficacy of stochastic quantization to enhance robustness and accuracy. Noise injection during quantization is explored to understand the impact of noise types and magnitudes on model performance. A comprehensive comparison between different applying scenarios for stochastic quantization and different noise types and magnitudes was implemented in this paper. Compared to the baseline deterministic quantization, chaotic quantization achieves a comparable accuracy, however, it achieves up to a 43% increase in accuracy against various attack scenarios. This highlights stochastic quantization as a promising defense mechanism. In addition, there is a crucial role played by the choice of noise type and magnitude in stochastic quantization. Lorenz and Henon noise distributions in stochastic quantization outperform traditional uniform and Gaussian noise in defending against attacks. A transferability analysis was discussed to understand the generalizability and effectiveness of the proposed stochastic quantization techniques. A cross-validation definition was newly evaluated in this scope to analyse the model’s stability and robustness against attacks. The study outperformed a quantization network technique and improved the model’s robustness and stability against adversarial attacks using chaotic quantization instead of deterministic quantization or even instead of stochastic quantization using traditional noise.

Discovery of sparse, reliable omic biomarkers with Stabl

Adoption of high-content omic technologies in clinical studies, coupled with computational methods, has yielded an abundance of candidate biomarkers. However, translating such findings into bona fide clinical biomarkers remains challenging. To facilitate this process, we introduce Stabl, a general machine learning method that identifies a sparse, reliable set of biomarkers by integrating noise injection and a data-driven signal-to-noise threshold into multivariable predictive modeling. Evaluation of Stabl on synthetic datasets and five independent clinical studies demonstrates improved biomarker sparsity and reliability compared to commonly used sparsity-promoting regularization methods while maintaining predictive performance; it distills datasets containing 1,400–35,000 features down to 4–34 candidate biomarkers. Stabl extends to multi-omic integration tasks, enabling biological interpretation of complex predictive models, as it hones in on a shortlist of proteomic, metabolomic and cytometric events predicting labor onset, microbial biomarkers of pre-term birth and a pre-operative immune signature of post-surgical infections. Stabl is available at https://github.com/gregbellan/Stabl.

A Hybrid Regularized Multilayer Perceptron for Input Noise Immunity

The immunity of multilayer perceptron (MLP) is less effective towards input noise. In this article, we have focused on the robustness of MLP with respect to input noise where noise can be additive or multiplicative. Here, we have proposed a DropConnect based regularized MLP to reduce the coadaptation among the neurons of the hidden layer. At first, we have empirically and statistically shown that by reducing coadaptation among the hidden neurons, an MLP can achieve better noise immunity. We have also empirically shown that an MLP with input noise injection and <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$l_{2}$</tex-math></inline-formula> regularizer is an effective approach to improve its noise immunity. However, the results indicate that it does not adjust the coadaptation among the hidden neurons. Therefore, for further improvement, we have proposed a hybrid regularized MLP (HRMLP), where DropConnect is combined with the noise injection and <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$l_{2}$</tex-math></inline-formula> regularizer. In addition to input noise, we have also verified the robustness of HRMLP with respect to <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\mathbf {20\%}$</tex-math></inline-formula> outliers in the data set. To justify the effectiveness of the proposed HRMLP, we have compared it with MLP, MLP with noise injection, MLP with <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$l_{2}$</tex-math></inline-formula> regularizer, MLP with noise injection and <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$l_{2}$</tex-math></inline-formula> regularizer, and MLP with DropConnect along with two state-of-the-art works based on <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\mathbf {20}$</tex-math></inline-formula> standard data sets. The experimental results for both noisy inputs and outliers confirm that the performance of HRMLP is significant compared to other methods.

Automated Anomaly Detection for Surface Defects by Dual Generative Networks With Limited Training Data

The manufacturing process in the pharmaceutical industry requires for using vessels or tanks which inner surfaces can undergo fouling and/or may develop visible defects that have to be cleaned away. Surface contamination can present itself in different forms, and there is no available image database of stains. In this paper, we propose to use an unsupervised anomaly detection approach which trains models on one or just a few images of clean surfaces. First, we propose an inpainting GAN with global and local generators to make full use of every single image and to reduce the overfitting caused by a limited number of images in the training dataset. Furthermore, we propose a periodic noise injection technique to increase the number of images for training and to improve the detection performance of the network. The experimental results demonstrate that the proposed network and noise injection technique achieve significant performance gains in the challenging task of anomaly detection on the surfaces of pharmaceutical equipment and other real-world industrial datasets. We also consider possible generalizations of the proposed method.

Hierarchical Perceptual Noise Injection for Social Media Fingerprint Privacy Protection.

Billions of people share images from their daily lives on social media every day. However, their biometric information (e.g., fingerprints) could be easily stolen from these images. The threat of fingerprint leakage from social media has created a strong desire to anonymize shared images while maintaining image quality, since fingerprints act as a lifelong individual biometric password. To guard the fingerprint leakage, adversarial attack that involves adding imperceptible perturbations to fingerprint images have emerged as a feasible solution. However, existing works of this kind are either weak in black-box transferability or cause the images to have an unnatural appearance. Motivated by the visual perception hierarchy (i.e., high-level perception exploits model-shared semantics that transfer well across models while low-level perception extracts primitive stimuli that result in high visual sensitivity when a suspicious stimulus is provided), we propose FingerSafe, a hierarchical perceptual protective noise injection framework to address the above mentioned problems. For black-box transferability, we inject protective noises into the fingerprint orientation field to perturb the model-shared high-level semantics (i.e., fingerprint ridges). Considering visual naturalness, we suppress the low-level local contrast stimulus by regularizing the response of the Lateral Geniculate Nucleus. Our proposed FingerSafe is the first to provide feasible fingerprint protection in both digital (up to 94.12%) and realistic scenarios (Twitter and Facebook, up to 68.75%). Our code can be found at https://github.com/nlsde-safety-team/FingerSafe.

Noise-induced chaos in an electronic Burridge-Knopoff model analog

Electronic analog circuits are valuable platforms to experimentally address physical systems whose implementation would otherwise be unwieldy. Besides validating numerical predictions, electronic analogs of nonlinear systems are also ideal tools to study the effect of noise injection in realistic conditions and are often more scalable than numerical approaches. In this work, we address the effect of noise injection in an electronic analog of the Burridge-Knopoff mechanical model of earthquake faults. Noise is injected into the friction term of the mechanical system: interestingly, the presence of a sufficient level of noise can trigger the onset of chaos in otherwise periodic regimes.

Remaining Useful Life Estimation based on Noise Injection and a Kalman Filter Ensemble of modified Bagging Predictors

Remaining Useful Life Estimation based on Noise Injection and a Kalman Filter Ensemble of modified Bagging Predictors

Implementation of Captcha Mechanisms using Deep Learning to Prevent Automated Bot Attacks

Online platforms' integrity and security are seriously threatened by the growth of automated bot attacks, necessitating the development of effective methods for telling harmful bots apart from legitimate users. In order to successfully combat automated bot attacks, this project investigates the creation and application of CAPTCHA (Completely Automated Public Turing test to tell Computers and Humans Apart) methods utilising deep learning techniques. In this study, we employ deep learning to create and apply complex CAPTCHA-problems that help us distinguish between real users and automated bots. Convolutional neural networks and recurrent neural networks are employed to build dynamic, adjustable CAPTCHAs in order to keep up with the bots' evolving approaches. Our research focuses on creating CAPTCHAs that are simple to use for humans but difficult to understand by robots, which creates a favourable user experience for those who are actually human. Character skewing, background noise injection, and image obfuscation are some of the methods we use to safeguard our CAPTCHAs while maintaining their usability. Furthermore, we carry out exhaustive trials in the real world to assess the effectiveness of our methods based on in-depth CAPTCHA learning. We evaluate their resistance to a variety of attack methods, such as counterattacks and machine learning-based bot attacks, in order to confirm that they are resilient. The findings of our study show that utilising deep learning in CAPTCHA methods to thwart automated bot attacks is both feasible and effective. Our method makes the internet environment safer and more user-friendly while also enhancing security and reducing user annoyance. This work is an important step in the fight against the growing menace of automated bot attacks in the digital sphere.&#x0D;

Exploring the Efficacy of Base Data Augmentation Methods in Deep Learning-Based Radiograph Classification of Knee Joint Osteoarthritis

Diagnosing knee joint osteoarthritis (KOA), a major cause of disability worldwide, is challenging due to subtle radiographic indicators and the varied progression of the disease. Using deep learning for KOA diagnosis requires broad, comprehensive datasets. However, obtaining these datasets poses significant challenges due to patient privacy and data collection restrictions. Additive data augmentation, which enhances data variability, emerges as a promising solution. Yet, it’s unclear which augmentation techniques are most effective for KOA. Our study explored data augmentation methods, including adversarial techniques. We used strategies like horizontal cropping and region of interest (ROI) extraction, alongside adversarial methods such as noise injection and ROI removal. Interestingly, rotations improved performance, while methods like horizontal split were less effective. We discovered potential confounding regions using adversarial augmentation, shown in our models’ accurate classification of extreme KOA grades, even without the knee joint. This indicated a potential model bias towards irrelevant radiographic features. Removing the knee joint paradoxically increased accuracy in classifying early-stage KOA. Grad-CAM visualizations helped elucidate these effects. Our study contributed to the field by pinpointing augmentation techniques that either improve or impede model performance, in addition to recognizing potential confounding regions within radiographic images of knee osteoarthritis.