Abstract Federated learning (FL) is a distributed machine learning framework that coordinates clients to train models on their private datasets via a centralized server, thereby mitigating data privacy risks. However, the communication channels involved in this process are untrusted, leaving FL vulnerable to model poisoning attacks launched by adversaries through man-in-the-middle techniques. Such attacks can degrade the accuracy of the global model and ultimately cause the entire FL training process to fail. In this paper, we propose a defense mechanism that integrates secure verification with watermarking, with the primary goal of ensuring the integrity of models transmitted over communication channels and enabling highly reliable FL deployment. Our mechanism leverages a dual watermarking method: first, models are marked using specially generated samples, and then these samples are further watermarked based on a class histogram-inspired approach. This dual strategy enhances both model detection and watermark stealthiness. The key innovation of our method lies in its sensitivity to subtle tampering while imposing no loss in model accuracy. Experimental results demonstrate that our defense mechanism significantly strengthens the resilience of FL models against sophisticated model poisoning attacks, while maintaining high accuracy and reliability.

Backdoor attacks pose a significant threat during the model's training phase. Attackers craft pre-defined triggers to break deep neural networks, ensuring the model accurately classifies clean samples during inference yet erroneously classifies samples added with these triggers. Recent studies have shown that speaker recognition systems trained on large-scale data are susceptible to backdoor attacks. Existing attackers employ unnoticed ambient sounds as triggers. However, these sounds are not inherently part of the training samples themselves. In essence, triggers can be designed to maintain an intrinsic connection with the original speech to enhance stealthiness. Our paper presents a novel attack methodology named Speed Master, which undermines deep neural networks by manipulating the speed of speech samples. Specifically, we execute poison-only backdoor attacks using speed or tempo adjustment. Changes in speech rate have become a common occurrence, as seen on platforms that allow users to adjust playback speed. In real-world scenarios, people naturally adjust their speaking rate depending on the context. As a result, changes in a speaker’s speech rate are typically perceived as normal and are unlikely to raise suspicion. Furthermore, detecting such subtle adjustments becomes challenging for users without reference speech. Our comprehensive experiments demonstrate that Speed Master can achieve an ASR over 99% in the digital domain, with only a 0.6% poisoning rate. Additionally, we validate the feasibility of Speed Master in the real world and its resistance to typical defensive measures.

Non-intrusive speech quality assessment (NISQA) is widely used in speech downstream tasks due to its ability to predict the quality of speech without a reference speech. However, few researchers have focused on the backdoor security of NISQA. Despite the backdoor defenses have been extensively studied to mitigate the threat of maliciously modifications in deep neural networks. In particular, semi-supervised based backdoor defenses have excellent defensive performance by depriving backdoor attacks of their most essential need. But these defense methods rely on data-augmentation consistency and thus cannot be applied to NISQA. In this work, we propose a backdoor defense based on semi-supervised and multi-task learning (SML). Semi-supervised learning is based on the simple assumption that the same input should be as consistent as possible in two similar models. Multi-task learning further improves the prediction performance of mean opinion score (MOS) by learning the tasks of perceptual evaluation of speech quality (PESQ), short-time objective intelligibility (STOI) and speech distortion index (SDI). Extensive experiments involving five backdoor defenses against five backdoor attacks on two benchmark datasets demonstrate the superiority of our SML approach.

Non-Intrusive speech quality assessment (NISQA) has gained significant attention for predicting speech’s mean opinion score (MOS) without requiring the reference speech. Researchers have gradually started to apply NISQA to various practical scenarios. However, little attention has been paid to the security of NISQA models. Backdoor attacks represent the most serious threat to deep neural networks (DNNs) due to the fact that backdoors possess a very high attack success rate once embedded. However, existing backdoor attacks assume that the attacker actively feeds samples containing triggers into the model during the inference phase. This is not adapted to the specific scenario of NISQA. And current backdoor attacks on regression tasks lack an objective metric to measure the attack performance. To address these issues, we propose a novel backdoor triggering approach (EventTrojan) that utilizes an event during the usage of the NISQA model as a trigger. Moreover, we innovatively provide an objective metric for backdoor attacks on regression tasks. Extensive experiments on four benchmark datasets demonstrate the effectiveness of the EventTrojan attack. Besides, it also has good resistance to several defense methods.

MOTIVATION: Identifying drug-target interactions (DTI) is crucial in drug discovery. Fragments are less complex and can accurately characterize local features, which is important in DTI prediction. Recently, deep learning (DL)-based methods predict DTI more efficiently. However, two challenges remain in existing DL-based methods: (i) some methods directly encode drugs and proteins into integers, ignoring the substructure representation; (ii) some methods learn the features of the drugs and proteins separately instead of considering their interactions. RESULTS: In this article, we propose a fragment-oriented method based on a multihead cross attention mechanism for predicting DTI, named FMCA-DTI. FMCA-DTI obtains multiple types of fragments of drugs and proteins by branch chain mining and category fragment mining. Importantly, FMCA-DTI utilizes the shared-weight-based multihead cross attention mechanism to learn the complex interaction features between different fragments. Experiments on three benchmark datasets show that FMCA-DTI achieves significantly improved performance by comparing it with four state-of-the-art baselines. AVAILABILITY AND IMPLEMENTATION: The code for this workflow is available at: https://github.com/jacky102022/FMCA-DTI.

Precisely forecasting the levels of [Formula: see text] is crucial for environmental conservation and human health. Thus, it serves as an essential indicator of atmospheric purity. In this paper, a [Formula: see text] concentration prediction model based on random forest and SHAP is proposed using air pollutants and meteorological conditions as the characterizing factors. Initially, pertinent information is gathered and subsequently manipulated, educated, and forecasted through the application of the random forest technique. Then, SHAP is used to explain the degree of influence of each feature in the model and the prediction results. Results of the experiment demonstrate that the random forest-based [Formula: see text] concentration prediction model for the three cities surpass the comparison model in the RMSE, MAE, and [Formula: see text] indicators. Examining SHAP values, the essential elements influencing the [Formula: see text] concentration are pinpointed.

In this paper, we propose the speech quality assessment transformer utilizing listener dependent modeling (SQAT-LD) mean opinion score (MOS) prediction system, which was submitted to the 2023 VoiceMOS Challenge. The system is based on a combination of self-supervised learning (SSL) models and listener-dependent modeling. Due to this challenge’s emphasis on real-world and challenging zero-shot out-of-domain MOS prediction in three different voice evaluation scenarios, we specifically designed a two-branch module to predict scores and weights for each frame, aiming to achieve better generalization. In the challenge, our system achieved fourth place in Track 1a, second place in Track 1b and first place in Track 2. Additionally, we conducted an ablation study to investigate the effectiveness of our proposed method.

In recent years, the business scale of my country’s circulation industry has continued to expand, and the output value has continued to increase. The leading role in guiding the transformation of the industrial economy has become more and more important. Based on this, this article discusses the research on the promotion of the Internet of Things technology to the circulation industry in the transformation of the resource‐based economy. The application of RFID technology and wireless sensor technology in the Internet of Things in the circulation industry can greatly improve work efficiency and information transmission accuracy. This article establishes the circulation industry based on the principles of science, system, safety, and relative independence. The evaluation index system analyzes the role of the circulation industry in the transformation of the resource‐based economy in terms of circulation scale, circulation structure, circulation efficiency, circulation innovation, etc., and uses the analytic hierarchy process and entropy method to analyze the collected data. With the support of RFID technology, the output value of the circulation industry in Province Y has reached 140.508 billion yuan in 2019, accounting for about 27% of the tertiary industry, and the number of employees in the circulation industry has also increased to 57.88%, which is a strong boost to the economy of Province Y. It has a greater contribution to the total economic volume. The research of this article has realized the economic transformation of resource‐based cities in the circulation industry and has a certain reference effect for the transformation and upgrading of similar cities.

This paper addresses the issue of color model learning and adaptation when color is used as a feature for object tracking in a dynamic scene. Under different environmental conditions, e.g. illumination changes or non-stationary scenes, a static color model is inadequate and color model adaptation is required. The color model adaptation for object tracking can be classified as an unsupervised (or semi-supervised) learning problem. The algorithm should be able to select the reliable training samples to update the color model automatically. A Bilateral Learning (BL) approach is proposed in this paper. The spatial and color information of the target are combined in the color model adaptation and color based object tracking procedure. The color model and spatial model are adapted in the color space and image space alternatively, which results in the color model adaptation and the localization of the target along the image sequence. Experimental results show the effectiveness and efficacy of the proposed method for the color model adaptation and object tracking under illumination changes and environmental noises.

The number of Chinese characters is very large, frequently it exceeds 5000 in daily usage. In order to achieve accuracy and speed in the recognition of handwritten Chinese characters, it is essential to have a well-organized model database. In the paper, the structural and statistical information of Chinese characters are represented by hierarchical attributed graphs. A heterogeneous multiway tree structure is used to organize the model database. For an input character, a corresponding model character in the database is found by a search process which can be divided into a number of simple and local decisions at different levels of the tree. The matching process becomes quite efficient and accurate.>