Abstract
This study harnesses state-of-the-art AI technology for chronic disease management, specifically in detecting various mental disorders through user-generated textual content. Existing studies typically rely on fully supervised machine learning, which presents challenges such as the labor-intensive manual process of annotating extensive training data for each disease and the need to design specialized deep learning architectures for each problem. To address such challenges, we propose a novel framework that leverages advanced AI techniques, including large language models and multi-prompt engineering. Specifically, we address two key technical challenges in data-driven chronic disease management: (1) developing personalized prompts to represent each user's uniqueness and (2) incorporating medical knowledge into prompts to provide context for chronic disease detection, instruct learning objectives, and operationalize prediction goals. We evaluate our method using four mental disorders, which are prevalent chronic diseases worldwide, as research cases. On the depression detection task, our method (F1 = 0.975~0.978) significantly outperforms traditional supervised learning paradigms, including feature engineering (F1 = 0.760) and architecture engineering (F1 = 0.756). Meanwhile, our approach demonstrates success in few-shot learning, i.e., requiring only a minimal number of training examples to detect chronic diseases based on user-generated textual content (i.e., only 2, 10, or 100 subjects). Moreover, our method can be generalized to other mental disorder detection tasks, including anorexia, pathological gambling, and self-harm (F1 = 0.919~0.978).
SpacTor-T5: Pre-training T5 Models with Span Corruption and Replaced Token Detection
Authors: Authors: Ke Ye, Heinrich Jiang, Afshin Rostamizadeh, Ayan Chakrabarti, Giulia DeSalvo, Jean-François Kagy, Lazaros Karydas, Gui Citovsky, Sanjiv Kumar
Subjects: Machine Learning (cs.LG); Computation and Language (cs.CL)
Abstract
Pre-training large language models is known to be extremely resource intensive and often times inefficient, under-utilizing the information encapsulated in the training text sequences. In this paper, we present SpacTor, a new training procedure consisting of (1) a hybrid objective combining span corruption (SC) and token replacement detection (RTD), and (2) a two-stage curriculum that optimizes the hybrid objective over the initial $\tau$ iterations, then transitions to standard SC loss. We show empirically that the effectiveness of the hybrid objective is tied to the two-stage pre-training schedule, and provide extensive analysis on why this is the case. In our experiments with encoder-decoder architectures (T5) on a variety of NLP tasks, SpacTor-T5 yields the same downstream performance as standard SC pre-training, while enabling a 50% reduction in pre-training iterations and 40% reduction in total FLOPs. Alternatively, given the same amount of computing budget, we find that SpacTor results in significantly improved downstream benchmark performance.
A Repository-Level Dataset For Detecting, Classifying and Repairing Software Vulnerabilities
Abstract
Open-Source Software (OSS) vulnerabilities bring great challenges to the software security and pose potential risks to our society. Enormous efforts have been devoted into automated vulnerability detection, among which deep learning (DL)-based approaches have proven to be the most effective. However, the current labeled data present the following limitations: (1) \textbf{Tangled Patches}: Developers may submit code changes unrelated to vulnerability fixes within patches, leading to tangled patches. (2) \textbf{Lacking Inter-procedural Vulnerabilities}: The existing vulnerability datasets typically contain function-level and file-level vulnerabilities, ignoring the relations between functions, thus rendering the approaches unable to detect the inter-procedural vulnerabilities. (3) \textbf{Outdated Patches}: The existing datasets usually contain outdated patches, which may bias the model during training. To address the above limitations, in this paper, we propose an automated data collection framework and construct the first repository-level high-quality vulnerability dataset named \textbf{ReposVul}. The proposed framework mainly contains three modules: (1) A vulnerability untangling module, aiming at distinguishing vulnerability-fixing related code changes from tangled patches, in which the Large Language Models (LLMs) and static analysis tools are jointly employed. (2) A multi-granularity dependency extraction module, aiming at capturing the inter-procedural call relationships of vulnerabilities, in which we construct multiple-granularity information for each vulnerability patch, including repository-level, file-level, function-level, and line-level. (3) A trace-based filtering module, aiming at filtering the outdated patches, which leverages the file path trace-based filter and commit time trace-based filter to construct an up-to-date dataset.
Towards Multi-domain Face Landmark Detection with Synthetic Data from Diffusion model
Authors: Authors: Yuanming Li, Gwantae Kim, Jeong-gi Kwak, Bon-hwa Ku, Hanseok Ko
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
Recently, deep learning-based facial landmark detection for in-the-wild faces has achieved significant improvement. However, there are still challenges in face landmark detection in other domains (e.g. cartoon, caricature, etc). This is due to the scarcity of extensively annotated training data. To tackle this concern, we design a two-stage training approach that effectively leverages limited datasets and the pre-trained diffusion model to obtain aligned pairs of landmarks and face in multiple domains. In the first stage, we train a landmark-conditioned face generation model on a large dataset of real faces. In the second stage, we fine-tune the above model on a small dataset of image-landmark pairs with text prompts for controlling the domain. Our new designs enable our method to generate high-quality synthetic paired datasets from multiple domains while preserving the alignment between landmarks and facial features. Finally, we fine-tuned a pre-trained face landmark detection model on the synthetic dataset to achieve multi-domain face landmark detection. Our qualitative and quantitative results demonstrate that our method outperforms existing methods on multi-domain face landmark detection.
Multitask Active Learning for Graph Anomaly Detection
Authors: Authors: Wenjing Chang, Kay Liu, Kaize Ding, Philip S. Yu, Jianjun Yu
Subjects: Machine Learning (cs.LG); Social and Information Networks (cs.SI)
Abstract
In the web era, graph machine learning has been widely used on ubiquitous graph-structured data. As a pivotal component for bolstering web security and enhancing the robustness of graph-based applications, the significance of graph anomaly detection is continually increasing. While Graph Neural Networks (GNNs) have demonstrated efficacy in supervised and semi-supervised graph anomaly detection, their performance is contingent upon the availability of sufficient ground truth labels. The labor-intensive nature of identifying anomalies from complex graph structures poses a significant challenge in real-world applications. Despite that, the indirect supervision signals from other tasks (e.g., node classification) are relatively abundant. In this paper, we propose a novel MultItask acTIve Graph Anomaly deTEction framework, namely MITIGATE. Firstly, by coupling node classification tasks, MITIGATE obtains the capability to detect out-of-distribution nodes without known anomalies. Secondly, MITIGATE quantifies the informativeness of nodes by the confidence difference across tasks, allowing samples with conflicting predictions to provide informative yet not excessively challenging information for subsequent training. Finally, to enhance the likelihood of selecting representative nodes that are distant from known patterns, MITIGATE adopts a masked aggregation mechanism for distance measurement, considering both inherent features of nodes and current labeled status. Empirical studies on four datasets demonstrate that MITIGATE significantly outperforms the state-of-the-art methods for anomaly detection. Our code is publicly available at: https://github.com/AhaChang/MITIGATE.
AMANet: Advancing SAR Ship Detection with Adaptive Multi-Hierarchical Attention Network
Abstract
Recently, methods based on deep learning have been successfully applied to ship detection for synthetic aperture radar (SAR) images. Despite the development of numerous ship detection methodologies, detecting small and coastal ships remains a significant challenge due to the limited features and clutter in coastal environments. For that, a novel adaptive multi-hierarchical attention module (AMAM) is proposed to learn multi-scale features and adaptively aggregate salient features from various feature layers, even in complex environments. Specifically, we first fuse information from adjacent feature layers to enhance the detection of smaller targets, thereby achieving multi-scale feature enhancement. Then, to filter out the adverse effects of complex backgrounds, we dissect the previously fused multi-level features on the channel, individually excavate the salient regions, and adaptively amalgamate features originating from different channels. Thirdly, we present a novel adaptive multi-hierarchical attention network (AMANet) by embedding the AMAM between the backbone network and the feature pyramid network (FPN). Besides, the AMAM can be readily inserted between different frameworks to improve object detection. Lastly, extensive experiments on two large-scale SAR ship detection datasets demonstrate that our AMANet method is superior to state-of-the-art methods.
MF-AED-AEC: Speech Emotion Recognition by Leveraging Multimodal Fusion, ASR Error Detection, and ASR Error Correction
Authors: Authors: Jiajun He, Xiaohan Shi, Xingfeng Li, Tomoki Toda
Subjects: Computation and Language (cs.CL); Multimedia (cs.MM); Sound (cs.SD); Audio and Speech Processing (eess.AS)
Abstract
The prevalent approach in speech emotion recognition (SER) involves integrating both audio and textual information to comprehensively identify the speaker's emotion, with the text generally obtained through automatic speech recognition (ASR). An essential issue of this approach is that ASR errors from the text modality can worsen the performance of SER. Previous studies have proposed using an auxiliary ASR error detection task to adaptively assign weights of each word in ASR hypotheses. However, this approach has limited improvement potential because it does not address the coherence of semantic information in the text. Additionally, the inherent heterogeneity of different modalities leads to distribution gaps between their representations, making their fusion challenging. Therefore, in this paper, we incorporate two auxiliary tasks, ASR error detection (AED) and ASR error correction (AEC), to enhance the semantic coherence of ASR text, and further introduce a novel multi-modal fusion (MF) method to learn shared representations across modalities. We refer to our method as MF-AED-AEC. Experimental results indicate that MF-AED-AEC significantly outperforms the baseline model by a margin of 4.1\%.
Abstract
Recently,the detection transformer has gained substantial attention for its inherent minimal post-processing requirement.However,this paradigm relies on abundant training data,yet in the context of the cross-domain adaptation,insufficient labels in the target domain exacerbate issues of class imbalance and model performance degradation.To address these challenges, we propose a novel class-aware cross domain detection transformer based on the adversarial learning and mean-teacher framework.First,considering the inconsistencies between the classification and regression tasks,we introduce an IoU-aware prediction branch and exploit the consistency of classification and location scores to filter and reweight pseudo labels.Second, we devise a dynamic category threshold refinement to adaptively manage model confidence.Third,to alleviate the class imbalance,an instance-level class-aware contrastive learning module is presented to encourage the generation of discriminative features for each class,particularly benefiting minority classes.Experimental results across diverse domain-adaptive scenarios validate our method's effectiveness in improving performance and alleviating class imbalance issues,which outperforms the state-of-the-art transformer based methods.
DDI-CoCo: A Dataset For Understanding The Effect Of Color Contrast In Machine-Assisted Skin Disease Detection
Abstract
Skin tone as a demographic bias and inconsistent human labeling poses challenges in dermatology AI. We take another angle to investigate color contrast's impact, beyond skin tones, on malignancy detection in skin disease datasets: We hypothesize that in addition to skin tones, the color difference between the lesion area and skin also plays a role in malignancy detection performance of dermatology AI models. To study this, we first propose a robust labeling method to quantify color contrast scores of each image and validate our method by showing small labeling variations. More importantly, applying our method to \textit{the only} diverse-skin tone and pathologically-confirmed skin disease dataset DDI, yields \textbf{DDI-CoCo Dataset}, and we observe a performance gap between the high and low color difference groups. This disparity remains consistent across various state-of-the-art (SoTA) image classification models, which supports our hypothesis. Furthermore, we study the interaction between skin tone and color difference effects and suggest that color difference can be an additional reason behind model performance bias between skin tones. Our work provides a complementary angle to dermatology AI for improving skin disease detection.
Towards Explainable Harmful Meme Detection through Multimodal Debate between Large Language Models
Authors: Authors: Hongzhan Lin, Ziyang Luo, Wei Gao, Jing Ma, Bo Wang, Ruichao Yang
Subjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
Abstract
The age of social media is flooded with Internet memes, necessitating a clear grasp and effective identification of harmful ones. This task presents a significant challenge due to the implicit meaning embedded in memes, which is not explicitly conveyed through the surface text and image. However, existing harmful meme detection methods do not present readable explanations that unveil such implicit meaning to support their detection decisions. In this paper, we propose an explainable approach to detect harmful memes, achieved through reasoning over conflicting rationales from both harmless and harmful positions. Specifically, inspired by the powerful capacity of Large Language Models (LLMs) on text generation and reasoning, we first elicit multimodal debate between LLMs to generate the explanations derived from the contradictory arguments. Then we propose to fine-tune a small language model as the debate judge for harmfulness inference, to facilitate multimodal fusion between the harmfulness rationales and the intrinsic multimodal information within memes. In this way, our model is empowered to perform dialectical reasoning over intricate and implicit harm-indicative patterns, utilizing multimodal explanations originating from both harmless and harmful arguments. Extensive experiments on three public meme datasets demonstrate that our harmful meme detection approach achieves much better performance than state-of-the-art methods and exhibits a superior capacity for explaining the meme harmfulness of the model predictions.
Classification of Radiologically Isolated Syndrome and Clinically Isolated Syndrome with Machine-Learning Techniques
Authors: Authors: V Mato-Abad, A Labiano-Fontcuberta, S Rodriguez-Yanez, R Garcia-Vazquez, CR Munteanu, J Andrade-Garda, A Domingo-Santos, V Galan Sanchez-Seco, Y Aladro, ML Martinez-Gines, L Ayuso, J Benito-Leon
Abstract
Background and purpose: The unanticipated detection by magnetic resonance imaging (MRI) in the brain of asymptomatic subjects of white matter lesions suggestive of multiple sclerosis (MS) has been named radiologically isolated syndrome (RIS). As the difference between early MS [i.e. clinically isolated syndrome (CIS)] and RIS is the occurrence of a clinical event, it is logical to improve detection of the subclinical form without interfering with MRI as there are radiological diagnostic criteria for that. Our objective was to use machine-learning classification methods to identify morphometric measures that help to discriminate patients with RIS from those with CIS. Methods: We used a multimodal 3-T MRI approach by combining MRI biomarkers (cortical thickness, cortical and subcortical grey matter volume, and white matter integrity) of a cohort of 17 patients with RIS and 17 patients with CIS for single-subject level classification. Results: The best proposed models to predict the diagnosis of CIS and RIS were based on the Naive Bayes, Bagging and Multilayer Perceptron classifiers using only three features: the left rostral middle frontal gyrus volume and the fractional anisotropy values in the right amygdala and right lingual gyrus. The Naive Bayes obtained the highest accuracy [overall classification, 0.765; area under the receiver operating characteristic (AUROC), 0.782]. Conclusions: A machine-learning approach applied to multimodal MRI data may differentiate between the earliest clinical expressions of MS (CIS and RIS) with an accuracy of 78%. Keywords: Bagging; Multilayer Perceptron; Naive Bayes classifier; clinically isolated syndrome; diffusion tensor imaging; machine-learning; magnetic resonance imaging; multiple sclerosis; radiologically isolated syndrome.
Generating Synthetic Health Sensor Data for Privacy-Preserving Wearable Stress Detection
Authors: Authors: Lucas Lange, Nils Wenzlitschke, Erhard Rahm
Subjects: Machine Learning (cs.LG); Cryptography and Security (cs.CR)
Abstract
Smartwatch health sensor data is increasingly utilized in smart health applications and patient monitoring, including stress detection. However, such medical data often comprises sensitive personal information and is resource-intensive to acquire for research purposes. In response to this challenge, we introduce the privacy-aware synthetization of multi-sensor smartwatch health readings related to moments of stress. Our method involves the generation of synthetic sequence data through Generative Adversarial Networks (GANs), coupled with the implementation of Differential Privacy (DP) safeguards for protecting patient information during model training. To ensure the integrity of our synthetic data, we employ a range of quality assessments and monitor the plausibility between synthetic and original data. To test the usefulness, we create private machine learning models on a commonly used, albeit small, stress detection dataset, exploring strategies for enhancing the existing data foundation with our synthetic data. Through our GAN-based augmentation methods, we observe improvements in model performance, both in non-private (0.45% F1) and private (11.90-15.48% F1) training scenarios. We underline the potential of differentially private synthetic data in optimizing utility-privacy trade-offs, especially with limited availability of real training samples.
Past, Present, Future: A Comprehensive Exploration of AI Use Cases in the UMBRELLA IoT Testbed
Authors: Authors: Peizheng Li, Ioannis Mavromatis, Aftab Khan
Subjects: Networking and Internet Architecture (cs.NI); Artificial Intelligence (cs.AI)
Abstract
UMBRELLA is a large-scale, open-access Internet of Things (IoT) ecosystem incorporating over 200 multi-sensor multi-wireless nodes, 20 collaborative robots, and edge-intelligence-enabled devices. This paper provides a guide to the implemented and prospective artificial intelligence (AI) capabilities of UMBRELLA in real-world IoT systems. Four existing UMBRELLA applications are presented in detail: 1) An automated streetlight monitoring for detecting issues and triggering maintenance alerts; 2) A Digital twin of building environments providing enhanced air quality sensing with reduced cost; 3) A large-scale Federated Learning framework for reducing communication overhead; and 4) An intrusion detection for containerised applications identifying malicious activities. Additionally, the potential of UMBRELLA is outlined for future smart city and multi-robot crowdsensing applications enhanced by semantic communications and multi-agent planning. Finally, to realise the above use-cases we discuss the need for a tailored MLOps platform to automate UMBRELLA model pipelines and establish trust.
Text Categorization Can Enhance Domain-Agnostic Stopword Extraction
Authors: Authors: Houcemeddine Turki, Naome A. Etori, Mohamed Ali Hadj Taieb, Abdul-Hakeem Omotayo, Chris Chinenye Emezue, Mohamed Ben Aouicha, Ayodele Awokoya, Falalu Ibrahim Lawan, Doreen Nixdorf
Subjects: Computation and Language (cs.CL); Machine Learning (cs.LG)
Abstract
This paper investigates the role of text categorization in streamlining stopword extraction in natural language processing (NLP), specifically focusing on nine African languages alongside French. By leveraging the MasakhaNEWS, African Stopwords Project, and MasakhaPOS datasets, our findings emphasize that text categorization effectively identifies domain-agnostic stopwords with over 80% detection success rate for most examined languages. Nevertheless, linguistic variances result in lower detection rates for certain languages. Interestingly, we find that while over 40% of stopwords are common across news categories, less than 15% are unique to a single category. Uncommon stopwords add depth to text but their classification as stopwords depends on context. Therefore combining statistical and linguistic approaches creates comprehensive stopword lists, highlighting the value of our hybrid method. This research enhances NLP for African languages and underscores the importance of text categorization in stopword extraction.
Finite-Precision Arithmetic Transceiver for Massive MIMO Systems
Authors: Authors: Yiming Fang, Li Chen, Yunfei Chen, Huarui Yin
Subjects: Information Theory (cs.IT); Signal Processing (eess.SP)
Abstract
Efficient implementation of massive multiple-input-multiple-output (MIMO) transceivers is essential for the next-generation wireless networks. To reduce the high computational complexity of the massive MIMO transceiver, in this paper, we propose a new massive MIMO architecture using finite-precision arithmetic. First, we conduct the rounding error analysis and derive the lower bound of the achievable rate for single-input-multiple-output (SIMO) using maximal ratio combining (MRC) and multiple-input-single-output (MISO) systems using maximal ratio transmission (MRT) with finite-precision arithmetic. Then, considering the multi-user scenario, the rounding error analysis of zero-forcing (ZF) detection and precoding is derived by using the normal equations (NE) method. The corresponding lower bounds of the achievable sum rate are also derived and asymptotic analyses are presented. Built upon insights from these analyses and lower bounds, we propose a mixed-precision architecture for massive MIMO systems to offset performance gaps due to finite-precision arithmetic. The corresponding analysis of rounding errors and computational costs is obtained. Simulation results validate the derived bounds and underscore the superiority of the proposed mixed-precision architecture to the conventional structure.
Faster Combinatorial k-Clique Algorithms
Authors: Authors: Amir Abboud, Nick Fischer, Yarin Shechter
Abstract
Detecting if a graph contains a $k$-Clique is one of the most fundamental problems in computer science. The asymptotically fastest algorithm runs in time $O(n^{\omega k/3})$, where $\omega$ is the exponent of Boolean matrix multiplication. To date, this is the only technique capable of beating the trivial $O(n^k)$ bound by a polynomial factor. Due to this technique's various limitations, much effort has gone into designing "combinatorial" algorithms that improve over exhaustive search via other techniques. The first contribution of this work is a faster combinatorial algorithm for $k$-Clique, improving Vassilevska's bound of $O(n^{k}/\log^{k-1}{n})$ by two log factors. Technically, our main result is a new reduction from $k$-Clique to Triangle detection that exploits the same divide-and-conquer at the core of recent combinatorial algorithms by Chan (SODA'15) and Yu (ICALP'15). Our second contribution is exploiting combinatorial techniques to improve the state-of-the-art (even of non-combinatorial algorithms) for generalizations of the $k$-Clique problem. In particular, we give the first $o(n^k)$ algorithm for $k$-clique in hypergraphs and an $O(n^3/\log^{2.25}{n} + t)$ algorithm for listing $t$ triangles in a graph.
Research about the Ability of LLM in the Tamper-Detection Area
Authors: Authors: Xinyu Yang, Jizhe Zhou
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
In recent years, particularly since the early 2020s, Large Language Models (LLMs) have emerged as the most powerful AI tools in addressing a diverse range of challenges, from natural language processing to complex problem-solving in various domains. In the field of tamper detection, LLMs are capable of identifying basic tampering activities.To assess the capabilities of LLMs in more specialized domains, we have collected five different LLMs developed by various companies: GPT-4, LLaMA, Bard, ERNIE Bot 4.0, and Tongyi Qianwen. This diverse range of models allows for a comprehensive evaluation of their performance in detecting sophisticated tampering instances.We devised two domains of detection: AI-Generated Content (AIGC) detection and manipulation detection. AIGC detection aims to test the ability to distinguish whether an image is real or AI-generated. Manipulation detection, on the other hand, focuses on identifying tampered images. According to our experiments, most LLMs can identify composite pictures that are inconsistent with logic, and only more powerful LLMs can distinguish logical, but visible signs of tampering to the human eye. All of the LLMs can't identify carefully forged images and very realistic images generated by AI. In the area of tamper detection, LLMs still have a long way to go, particularly in reliably identifying highly sophisticated forgeries and AI-generated images that closely mimic reality.
Delocate: Detection and Localization for Deepfake Videos with Randomly-Located Tampered Traces
Authors: Authors: Juan Hu, Xin Liao, Difei Gao, Satoshi Tsutsui, Qian Wang, Zheng Qin, Mike Zheng Shou
Subjects: Computer Vision and Pattern Recognition (cs.CV); Cryptography and Security (cs.CR)
Abstract
Deepfake videos are becoming increasingly realistic, showing subtle tampering traces on facial areasthat vary between frames. Consequently, many existing Deepfake detection methods struggle to detect unknown domain Deepfake videos while accurately locating the tampered region. To address thislimitation, we propose Delocate, a novel Deepfake detection model that can both recognize andlocalize unknown domain Deepfake videos. Ourmethod consists of two stages named recoveringand localization. In the recovering stage, the modelrandomly masks regions of interest (ROIs) and reconstructs real faces without tampering traces, resulting in a relatively good recovery effect for realfaces and a poor recovery effect for fake faces. Inthe localization stage, the output of the recoveryphase and the forgery ground truth mask serve assupervision to guide the forgery localization process. This process strategically emphasizes the recovery phase of fake faces with poor recovery, facilitating the localization of tampered regions. Ourextensive experiments on four widely used benchmark datasets demonstrate that Delocate not onlyexcels in localizing tampered areas but also enhances cross-domain detection performance.
Addressing Data Quality Challenges in Observational Ambulatory Studies: Analysis, Methodologies and Practical Solutions for Wrist-worn Wearable Monitoring
Authors: Authors: Jonas Van Der Donckt, Nicolas Vandenbussche, Jeroen Van Der Donckt, Stephanie Chen, Marija Stojchevska, Mathias De Brouwer, Bram Steenwinckel, Koen Paemeleire, Femke Ongenae, Sofie Van Hoecke
Subjects: Computational Engineering, Finance, and Science (cs.CE)
Abstract
Chronic disease management and follow-up are vital for realizing sustained patient well-being and optimal health outcomes. Recent advancements in wearable sensing technologies, particularly wrist-worn devices, offer promising solutions for longitudinal patient follow-up by shifting from subjective, intermittent self-reporting to objective, continuous monitoring. However, collecting and analyzing wearable data presents unique challenges, such as data entry errors, non-wear periods, missing wearable data, and wearable artifacts. We therefore present an in-depth exploration of data analysis challenges tied to wrist-worn wearables and ambulatory label acquisition, using two real-world datasets (i.e., mBrain21 and ETRI lifelog2020). We introduce novel practical countermeasures, including participant compliance visualizations, interaction-triggered questionnaires to assess personal bias, and an optimized wearable non-wear detection pipeline. Further, we propose a visual analytics approach to validate processing pipelines using scalable tools such as tsflex and Plotly-Resampler. Lastly, we investigate the impact of missing wearable data on "window-of-interest" analysis methodologies. Prioritizing transparency and reproducibility, we offer open access to our detailed code examples, facilitating adaptation in future wearable research. In conclusion, our contributions provide actionable approaches for wearable data collection and analysis in chronic disease management.
Interleaving One-Class and Weakly-Supervised Models with Adaptive Thresholding for Unsupervised Video Anomaly Detection
Authors: Authors: Yongwei Nie, Hao Huang, Chengjiang Long, Qing Zhang, Pradipta Maji, Hongmin Cai
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
Without human annotations, a typical Unsupervised Video Anomaly Detection (UVAD) method needs to train two models that generate pseudo labels for each other. In previous work, the two models are closely entangled with each other, and it is not known how to upgrade their method without modifying their training framework significantly. Second, previous work usually adopts fixed thresholding to obtain pseudo labels, however the user-specified threshold is not reliable which inevitably introduces errors into the training process. To alleviate these two problems, we propose a novel interleaved framework that alternately trains a One-Class Classification (OCC) model and a Weakly-Supervised (WS) model for UVAD. The OCC or WS models in our method can be easily replaced with other OCC or WS models, which facilitates our method to upgrade with the most recent developments in both fields. For handling the fixed thresholding problem, we break through the conventional cognitive boundary and propose a weighted OCC model that can be trained on both normal and abnormal data. We also propose an adaptive mechanism for automatically finding the optimal threshold for the WS model in a loose to strict manner. Experiments demonstrate that the proposed UVAD method outperforms previous approaches.
PanAf20K: A Large Video Dataset for Wild Ape Detection and Behaviour Recognition
Authors: Authors: Otto Brookes, Majid Mirmehdi, Colleen Stephens, Samuel Angedakin, Katherine Corogenes, Dervla Dowd, Paula Dieguez, Thurston C. Hicks, Sorrel Jones, Kevin Lee, Vera Leinert, Juan Lapuente, Maureen S. McCarthy, Amelia Meier, Mizuki Murai, Emmanuelle Normand, Virginie Vergnes, Erin G. Wessling, Roman M. Wittig, Kevin Langergraber, Nuria Maldonado, Xinyu Yang, Klaus Zuberbuhler, Christophe Boesch, Mimi Arandjelovic, Hjalmar Kuhl, Tilo Burghardt
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
We present the PanAf20K dataset, the largest and most diverse open-access annotated video dataset of great apes in their natural environment. It comprises more than 7 million frames across ~20,000 camera trap videos of chimpanzees and gorillas collected at 18 field sites in tropical Africa as part of the Pan African Programme: The Cultured Chimpanzee. The footage is accompanied by a rich set of annotations and benchmarks making it suitable for training and testing a variety of challenging and ecologically important computer vision tasks including ape detection and behaviour recognition. Furthering AI analysis of camera trap information is critical given the International Union for Conservation of Nature now lists all species in the great ape family as either Endangered or Critically Endangered. We hope the dataset can form a solid basis for engagement of the AI community to improve performance, efficiency, and result interpretation in order to support assessments of great ape presence, abundance, distribution, and behaviour and thereby aid conservation efforts.
RIS Empowered Near-Field Covert Communications
Authors: Authors: Jun Liu, Gang Yang, Yuanwei Liu, Xiangyun Zhou
Subjects: Information Theory (cs.IT); Signal Processing (eess.SP)
Abstract
This paper studies an extremely large-scale reconfigurable intelligent surface (XL-RIS) empowered covert communication system in the near-field region. Alice covertly transmits messages to Bob with the assistance of the XL-RIS, while evading detection by Willie. To enhance the covert communication performance, we maximize the achievable covert rate by jointly optimizing the hybrid analog and digital beamformers at Alice, as well as the reflection coefficient matrix at the XL-RIS. An alternating optimization algorithm is proposed to solve the joint beamforming design problem. For the hybrid beamformer design, a semi-closed-form solution for fully digital beamformer is first obtained by a weighted minimum mean-square error based algorithm, then the baseband digital and analog beamformers at Alice are designed by approximating the fully digital beamformer via manifold optimization. For the XL-RIS's reflection coefficient matrix design, a low-complexity alternating direction method of multipliers based algorithm is proposed to address the challenge of large-scale variables and unit-modulus constraints. Numerical results unveil that i) the near-field communications can achieve a higher covert rate than the far-field covert communications in general, and still realize covert transmission even if Willie is located at the same direction as Bob and closer to the XL-RIS; ii) the proposed algorithm can enhance the covert rate significantly compared to the benchmark schemes; iii) the proposed algorithm leads to a beam diffraction pattern that can bypass Willie and achieve high-rate covert transmission to Bob.
PLATE: A perception-latency aware estimator,
Authors: Authors: Rodrigo Aldana-López, Rosario Aragüés, Carlos Sagüés
Subjects: Systems and Control (eess.SY); Computer Vision and Pattern Recognition (cs.CV); Optimization and Control (math.OC)
Abstract
Target tracking is a popular problem with many potential applications. There has been a lot of effort on improving the quality of the detection of targets using cameras through different techniques. In general, with higher computational effort applied, i.e., a longer perception-latency, a better detection accuracy is obtained. However, it is not always useful to apply the longest perception-latency allowed, particularly when the environment doesn't require to and when the computational resources are shared between other tasks. In this work, we propose a new Perception-LATency aware Estimator (PLATE), which uses different perception configurations in different moments of time in order to optimize a certain performance measure. This measure takes into account a perception-latency and accuracy trade-off aiming for a good compromise between quality and resource usage. Compared to other heuristic frame-skipping techniques, PLATE comes with a formal complexity and optimality analysis. The advantages of PLATE are verified by several experiments including an evaluation over a standard benchmark with real data and using state of the art deep learning object detection methods for the perception stage.
Graph-Informed Neural Networks for Sparse Grid-Based Discontinuity Detectors
Authors: Authors: Francesco Della Santa, Sandra Pieraccini
Abstract
In this paper, we present a novel approach for detecting the discontinuity interfaces of a discontinuous function. This approach leverages Graph-Informed Neural Networks (GINNs) and sparse grids to address discontinuity detection also in domains of dimension larger than 3. GINNs, trained to identify troubled points on sparse grids, exploit graph structures built on the grids to achieve efficient and accurate discontinuity detection performances. We also introduce a recursive algorithm for general sparse grid-based detectors, characterized by convergence properties and easy applicability. Numerical experiments on functions with dimensions n = 2 and n = 4 demonstrate the efficiency and robust generalization of GINNs in detecting discontinuity interfaces. Notably, the trained GINNs offer portability and versatility, allowing integration into various algorithms and sharing among users.
Keyword: face recognition
Privacy-Preserving Face Recognition in Hybrid Frequency-Color Domain
Authors: Authors: Dong Han, Yong Li, Joachim Denzler
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
Face recognition technology has been deployed in various real-life applications. The most sophisticated deep learning-based face recognition systems rely on training millions of face images through complex deep neural networks to achieve high accuracy. It is quite common for clients to upload face images to the service provider in order to access the model inference. However, the face image is a type of sensitive biometric attribute tied to the identity information of each user. Directly exposing the raw face image to the service provider poses a threat to the user's privacy. Current privacy-preserving approaches to face recognition focus on either concealing visual information on model input or protecting model output face embedding. The noticeable drop in recognition accuracy is a pitfall for most methods. This paper proposes a hybrid frequency-color fusion approach to reduce the input dimensionality of face recognition in the frequency domain. Moreover, sparse color information is also introduced to alleviate significant accuracy degradation after adding differential privacy noise. Besides, an identity-specific embedding mapping scheme is applied to protect original face embedding by enlarging the distance among identities. Lastly, secure multiparty computation is implemented for safely computing the embedding distance during model inference. The proposed method performs well on multiple widely used verification datasets. Moreover, it has around 2.6% to 4.2% higher accuracy than the state-of-the-art in the 1:N verification scenario.
Keyword: augmentation
IndiText Boost: Text Augmentation for Low Resource India Languages
Abstract
Text Augmentation is an important task for low-resource languages. It helps deal with the problem of data scarcity. A data augmentation strategy is used to deal with the problem of data scarcity. Through the years, much work has been done on data augmentation for the English language. In contrast, very less work has been done on Indian languages. This is contrary to the fact that data augmentation is used to deal with data scarcity. In this work, we focus on implementing techniques like Easy Data Augmentation, Back Translation, Paraphrasing, Text Generation using LLMs, and Text Expansion using LLMs for text classification on different languages. We focus on 6 Indian languages namely: Sindhi, Marathi, Hindi, Gujarati, Telugu, and Sanskrit. According to our knowledge, no such work exists for text augmentation on Indian languages. We carry out binary as well as multi-class text classification to make our results more comparable. We get surprising results as basic data augmentation techniques surpass LLMs.
Catch-Up Mix: Catch-Up Class for Struggling Filters in CNN
Authors: Authors: Minsoo Kang, Minkoo Kang, Suhyun Kim
Subjects: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
Abstract
Deep learning has made significant advances in computer vision, particularly in image classification tasks. Despite their high accuracy on training data, deep learning models often face challenges related to complexity and overfitting. One notable concern is that the model often relies heavily on a limited subset of filters for making predictions. This dependency can result in compromised generalization and an increased vulnerability to minor variations. While regularization techniques like weight decay, dropout, and data augmentation are commonly used to address this issue, they may not directly tackle the reliance on specific filters. Our observations reveal that the heavy reliance problem gets severe when slow-learning filters are deprived of learning opportunities due to fast-learning filters. Drawing inspiration from image augmentation research that combats over-reliance on specific image regions by removing and replacing parts of images, our idea is to mitigate the problem of over-reliance on strong filters by substituting highly activated features. To this end, we present a novel method called Catch-up Mix, which provides learning opportunities to a wide range of filters during training, focusing on filters that may lag behind. By mixing activation maps with relatively lower norms, Catch-up Mix promotes the development of more diverse representations and reduces reliance on a small subset of filters. Experimental results demonstrate the superiority of our method in various vision classification datasets, providing enhanced robustness.
Generating Synthetic Health Sensor Data for Privacy-Preserving Wearable Stress Detection
Authors: Authors: Lucas Lange, Nils Wenzlitschke, Erhard Rahm
Subjects: Machine Learning (cs.LG); Cryptography and Security (cs.CR)
Abstract
Smartwatch health sensor data is increasingly utilized in smart health applications and patient monitoring, including stress detection. However, such medical data often comprises sensitive personal information and is resource-intensive to acquire for research purposes. In response to this challenge, we introduce the privacy-aware synthetization of multi-sensor smartwatch health readings related to moments of stress. Our method involves the generation of synthetic sequence data through Generative Adversarial Networks (GANs), coupled with the implementation of Differential Privacy (DP) safeguards for protecting patient information during model training. To ensure the integrity of our synthetic data, we employ a range of quality assessments and monitor the plausibility between synthetic and original data. To test the usefulness, we create private machine learning models on a commonly used, albeit small, stress detection dataset, exploring strategies for enhancing the existing data foundation with our synthetic data. Through our GAN-based augmentation methods, we observe improvements in model performance, both in non-private (0.45% F1) and private (11.90-15.48% F1) training scenarios. We underline the potential of differentially private synthetic data in optimizing utility-privacy trade-offs, especially with limited availability of real training samples.
Lessons on Datasets and Paradigms in Machine Learning for Symbolic Computation: A Case Study on CAD
Abstract
Symbolic Computation algorithms and their implementation in computer algebra systems often contain choices which do not affect the correctness of the output but can significantly impact the resources required: such choices can benefit from having them made separately for each problem via a machine learning model. This study reports lessons on such use of machine learning in symbolic computation, in particular on the importance of analysing datasets prior to machine learning and on the different machine learning paradigms that may be utilised. We present results for a particular case study, the selection of variable ordering for cylindrical algebraic decomposition, but expect that the lessons learned are applicable to other decisions in symbolic computation. We utilise an existing dataset of examples derived from applications which was found to be imbalanced with respect to the variable ordering decision. We introduce an augmentation technique for polynomial systems problems that allows us to balance and further augment the dataset, improving the machine learning results by 28\% and 38\% on average, respectively. We then demonstrate how the existing machine learning methodology used for the problem $-$ classification $-$ might be recast into the regression paradigm. While this does not have a radical change on the performance, it does widen the scope in which the methodology can be applied to make choices.
Do You Guys Want to Dance: Zero-Shot Compositional Human Dance Generation with Multiple Persons
Authors: Authors: Zhe Xu, Kun Wei, Xu Yang, Cheng Deng
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
Human dance generation (HDG) aims to synthesize realistic videos from images and sequences of driving poses. Despite great success, existing methods are limited to generating videos of a single person with specific backgrounds, while the generalizability for real-world scenarios with multiple persons and complex backgrounds remains unclear. To systematically measure the generalizability of HDG models, we introduce a new task, dataset, and evaluation protocol of compositional human dance generation (cHDG). Evaluating the state-of-the-art methods on cHDG, we empirically find that they fail to generalize to real-world scenarios. To tackle the issue, we propose a novel zero-shot framework, dubbed MultiDance-Zero, that can synthesize videos consistent with arbitrary multiple persons and background while precisely following the driving poses. Specifically, in contrast to straightforward DDIM or null-text inversion, we first present a pose-aware inversion method to obtain the noisy latent code and initialization text embeddings, which can accurately reconstruct the composed reference image. Since directly generating videos from them will lead to severe appearance inconsistency, we propose a compositional augmentation strategy to generate augmented images and utilize them to optimize a set of generalizable text embeddings. In addition, consistency-guided sampling is elaborated to encourage the background and keypoints of the estimated clean image at each reverse step to be close to those of the reference image, further improving the temporal consistency of generated videos. Extensive qualitative and quantitative results demonstrate the effectiveness and superiority of our approach.
Semi-Supervised Coupled Thin-Plate Spline Model for Rotation Correction and Beyond
Authors: Authors: Lang Nie, Chunyu Lin, Kang Liao, Shuaicheng Liu, Yao Zhao
Subjects: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
Abstract
Thin-plate spline (TPS) is a principal warp that allows for representing elastic, nonlinear transformation with control point motions. With the increase of control points, the warp becomes increasingly flexible but usually encounters a bottleneck caused by undesired issues, e.g., content distortion. In this paper, we explore generic applications of TPS in single-image-based warping tasks, such as rotation correction, rectangling, and portrait correction. To break this bottleneck, we propose the coupled thin-plate spline model (CoupledTPS), which iteratively couples multiple TPS with limited control points into a more flexible and powerful transformation. Concretely, we first design an iterative search to predict new control points according to the current latent condition. Then, we present the warping flow as a bridge for the coupling of different TPS transformations, effectively eliminating interpolation errors caused by multiple warps. Besides, in light of the laborious annotation cost, we develop a semi-supervised learning scheme to improve warping quality by exploiting unlabeled data. It is formulated through dual transformation between the searched control points of unlabeled data and its graphic augmentation, yielding an implicit correction consistency constraint. Finally, we collect massive unlabeled data to exhibit the benefit of our semi-supervised scheme in rotation correction. Extensive experiments demonstrate the superiority and universality of CoupledTPS over the existing state-of-the-art (SoTA) solutions for rotation correction and beyond. The code and data will be available at https://github.com/nie-lang/CoupledTPS.
LDCA: Local Descriptors with Contextual Augmentation for Few-Shot Learning
Authors: Authors: Maofa Wang, Bingchen Yan
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
Few-shot image classification has emerged as a key challenge in the field of computer vision, highlighting the capability to rapidly adapt to new tasks with minimal labeled data. Existing methods predominantly rely on image-level features or local descriptors, often overlooking the holistic context surrounding these descriptors. In this work, we introduce a novel approach termed "Local Descriptor with Contextual Augmentation (LDCA)". Specifically, this method bridges the gap between local and global understanding uniquely by leveraging an adaptive global contextual enhancement module. This module incorporates a visual transformer, endowing local descriptors with contextual awareness capabilities, ranging from broad global perspectives to intricate surrounding nuances. By doing so, LDCA transcends traditional descriptor-based approaches, ensuring each local feature is interpreted within its larger visual narrative. Extensive experiments underscore the efficacy of our method, showing a maximal absolute improvement of 20\% over the next-best on fine-grained classification datasets, thus demonstrating significant advancements in few-shot classification tasks.
Can GPT-3.5 Generate and Code Discharge Summaries?
Authors: Authors: Matúš Falis, Aryo Pradipta Gema, Hang Dong, Luke Daines, Siddharth Basetti, Michael Holder, Rose S Penfold, Alexandra Birch, Beatrice Alex
Abstract
Objective: To investigate GPT-3.5 in generating and coding medical documents with ICD-10 codes for data augmentation on low-resources labels. Materials and Methods: Employing GPT-3.5 we generated and coded 9,606 discharge summaries based on lists of ICD-10 code descriptions of patients with infrequent (generation) codes within the MIMIC-IV dataset. Combined with the baseline training set, this formed an augmented training set. Neural coding models were trained on baseline and augmented data and evaluated on a MIMIC-IV test set. We report micro- and macro-F1 scores on the full codeset, generation codes, and their families. Weak Hierarchical Confusion Matrices were employed to determine within-family and outside-of-family coding errors in the latter codesets. The coding performance of GPT-3.5 was evaluated both on prompt-guided self-generated data and real MIMIC-IV data. Clinical professionals evaluated the clinical acceptability of the generated documents. Results: Augmentation slightly hinders the overall performance of the models but improves performance for the generation candidate codes and their families, including one unseen in the baseline training data. Augmented models display lower out-of-family error rates. GPT-3.5 can identify ICD-10 codes by the prompted descriptions, but performs poorly on real data. Evaluators note the correctness of generated concepts while suffering in variety, supporting information, and narrative. Discussion and Conclusion: GPT-3.5 alone is unsuitable for ICD-10 coding. Augmentation positively affects generation code families but mainly benefits codes with existing examples. Augmentation reduces out-of-family errors. Discharge summaries generated by GPT-3.5 state prompted concepts correctly but lack variety, and authenticity in narratives. They are unsuitable for clinical practice.
Keyword: detection
Few-Shot Learning for Chronic Disease Management: Leveraging Large Language Models and Multi-Prompt Engineering with Medical Knowledge Injection
SpacTor-T5: Pre-training T5 Models with Span Corruption and Replaced Token Detection
A Repository-Level Dataset For Detecting, Classifying and Repairing Software Vulnerabilities
Towards Multi-domain Face Landmark Detection with Synthetic Data from Diffusion model
Multitask Active Learning for Graph Anomaly Detection
AMANet: Advancing SAR Ship Detection with Adaptive Multi-Hierarchical Attention Network
MF-AED-AEC: Speech Emotion Recognition by Leveraging Multimodal Fusion, ASR Error Detection, and ASR Error Correction
Enhancing cross-domain detection: adaptive class-aware contrastive transformer
DDI-CoCo: A Dataset For Understanding The Effect Of Color Contrast In Machine-Assisted Skin Disease Detection
Towards Explainable Harmful Meme Detection through Multimodal Debate between Large Language Models
Classification of Radiologically Isolated Syndrome and Clinically Isolated Syndrome with Machine-Learning Techniques
Generating Synthetic Health Sensor Data for Privacy-Preserving Wearable Stress Detection
Past, Present, Future: A Comprehensive Exploration of AI Use Cases in the UMBRELLA IoT Testbed
Text Categorization Can Enhance Domain-Agnostic Stopword Extraction
Finite-Precision Arithmetic Transceiver for Massive MIMO Systems
Faster Combinatorial k-Clique Algorithms
Research about the Ability of LLM in the Tamper-Detection Area
Delocate: Detection and Localization for Deepfake Videos with Randomly-Located Tampered Traces
Addressing Data Quality Challenges in Observational Ambulatory Studies: Analysis, Methodologies and Practical Solutions for Wrist-worn Wearable Monitoring
Interleaving One-Class and Weakly-Supervised Models with Adaptive Thresholding for Unsupervised Video Anomaly Detection
PanAf20K: A Large Video Dataset for Wild Ape Detection and Behaviour Recognition
RIS Empowered Near-Field Covert Communications
PLATE: A perception-latency aware estimator,
Graph-Informed Neural Networks for Sparse Grid-Based Discontinuity Detectors
Keyword: face recognition
Privacy-Preserving Face Recognition in Hybrid Frequency-Color Domain
Keyword: augmentation
IndiText Boost: Text Augmentation for Low Resource India Languages
Catch-Up Mix: Catch-Up Class for Struggling Filters in CNN
Generating Synthetic Health Sensor Data for Privacy-Preserving Wearable Stress Detection
Lessons on Datasets and Paradigms in Machine Learning for Symbolic Computation: A Case Study on CAD
Do You Guys Want to Dance: Zero-Shot Compositional Human Dance Generation with Multiple Persons
Semi-Supervised Coupled Thin-Plate Spline Model for Rotation Correction and Beyond
LDCA: Local Descriptors with Contextual Augmentation for Few-Shot Learning
Can GPT-3.5 Generate and Code Discharge Summaries?