Yifei Zhang, Dun Zeng, Jinglong Luo, Xinyu Fu, Guanzhong Chen, Zenglin Xu, Irwin King
Trustworthy Artificial Intelligence (TAI) has proven invaluable in curbing potential negative repercussions tied to AI applications. Within the TAI spectrum, Federated Learning (FL) emerges as a promising solution to safeguard personal information in distributed settings across a multitude of practical contexts. However, the realm of FL is not without its challenges. Especially worrisome are adversarial attacks targeting its algorithmic robustness and systemic confidentiality. Moreover, the presence of biases and opacity in prediction outcomes further complicates FL’s broader adoption. Consequently, there is a growing expectation for FL to instill trust. To address this, we chart out a comprehensive road-map for Trustworthy Federated Learning (TFL) and provide an overview of existing efforts across four pivotal dimensions: Privacy & Security , Robustness , Fairness , and Explainability . For each dimension, we identify potential pitfalls that might undermine TFL and present a curated selection of defensive strategies, enriched by a discourse on technical solutions tailored for TFL. Furthermore, we present potential challenges and future directions to be explored for in-depth TFL research with broader impacts.
{"title":"A Survey of Trustworthy Federated Learning: Issues, Solutions, and Challenges","authors":"Yifei Zhang, Dun Zeng, Jinglong Luo, Xinyu Fu, Guanzhong Chen, Zenglin Xu, Irwin King","doi":"10.1145/3678181","DOIUrl":"https://doi.org/10.1145/3678181","url":null,"abstract":"\u0000 Trustworthy Artificial Intelligence (TAI) has proven invaluable in curbing potential negative repercussions tied to AI applications. Within the TAI spectrum, Federated Learning (FL) emerges as a promising solution to safeguard personal information in distributed settings across a multitude of practical contexts. However, the realm of FL is not without its challenges. Especially worrisome are adversarial attacks targeting its algorithmic robustness and systemic confidentiality. Moreover, the presence of biases and opacity in prediction outcomes further complicates FL’s broader adoption. Consequently, there is a growing expectation for FL to instill trust. To address this, we chart out a comprehensive road-map for\u0000 Trustworthy Federated Learning (TFL)\u0000 and provide an overview of existing efforts across four pivotal dimensions:\u0000 Privacy & Security\u0000 ,\u0000 Robustness\u0000 ,\u0000 Fairness\u0000 , and\u0000 Explainability\u0000 . For each dimension, we identify potential pitfalls that might undermine TFL and present a curated selection of defensive strategies, enriched by a discourse on technical solutions tailored for TFL. Furthermore, we present potential challenges and future directions to be explored for in-depth TFL research with broader impacts.\u0000","PeriodicalId":48967,"journal":{"name":"ACM Transactions on Intelligent Systems and Technology","volume":null,"pages":null},"PeriodicalIF":7.2,"publicationDate":"2024-07-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141812628","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Thirasara Ariyarathna, Meisam Mohommady, Hye-young Paik, S. Kanhere
Decentralized machine learning, such as Federated Learning (FL), is widely adopted in many application domains. Especially in domains like recommendation systems, sharing gradients instead of private data has recently caught the research community’s attention. Personalized travel route recommendation utilizes users’ location data to recommend optimal travel routes. Location data is extremely privacy sensitive, presenting increased risks of exposing behavioural patterns and demographic attributes. FL for route recommendation can mitigate the sharing of location data. However, this paper shows that an adversary can recover the user trajectories used to train the federated recommendation models with high proximity accuracy. To this effect, we propose a novel attack called DeepSneak, which uses shared gradients obtained from global model training in FL to reconstruct private user trajectories. We formulate the attack as a regression problem and train a generative model by minimizing the distance between gradients. We validate the success of DeepSneak on two real-world trajectory datasets. The results show that we can recover the location trajectories of users with reasonable spatial and semantic accuracy.
{"title":"DeepSneak: User GPS Trajectory Reconstruction from Federated Route Recommendation Models","authors":"Thirasara Ariyarathna, Meisam Mohommady, Hye-young Paik, S. Kanhere","doi":"10.1145/3670412","DOIUrl":"https://doi.org/10.1145/3670412","url":null,"abstract":"Decentralized machine learning, such as Federated Learning (FL), is widely adopted in many application domains. Especially in domains like recommendation systems, sharing gradients instead of private data has recently caught the research community’s attention. Personalized travel route recommendation utilizes users’ location data to recommend optimal travel routes. Location data is extremely privacy sensitive, presenting increased risks of exposing behavioural patterns and demographic attributes. FL for route recommendation can mitigate the sharing of location data. However, this paper shows that an adversary can recover the user trajectories used to train the federated recommendation models with high proximity accuracy. To this effect, we propose a novel attack called DeepSneak, which uses shared gradients obtained from global model training in FL to reconstruct private user trajectories. We formulate the attack as a regression problem and train a generative model by minimizing the distance between gradients. We validate the success of DeepSneak on two real-world trajectory datasets. The results show that we can recover the location trajectories of users with reasonable spatial and semantic accuracy.","PeriodicalId":48967,"journal":{"name":"ACM Transactions on Intelligent Systems and Technology","volume":null,"pages":null},"PeriodicalIF":7.2,"publicationDate":"2024-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141815799","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In NLP (natural language processing), zero-shot topic classification requires machines to understand the contextual meanings of texts in a downstream task without using the corresponding labeled texts for training, which is highly desirable for various applications [2]. In this paper, we propose a novel approach to construct a zero-shot task-specific model called WC-SBERT with satisfactory performance. The proposed approach is highly efficient since it uses light self-training requiring target labels (target class names of downstream tasks) only, which is distinct from other research that uses both the target labels and the unlabeled texts for training. In particular, during the pre-training stage, WC-SBERT uses contrastive learning with the multiple negative ranking loss [9] to construct the pre-trained model based on the similarity between Wiki categories. For the self-training stage, online contrastive loss is utilized to reduce the distance between a target label and Wiki categories of similar Wiki pages to the label. Experimental results indicate that compared to existing self-training models, WC-SBERT achieves rapid inference on approximately 6.45 million Wiki text entries by utilizing pre-stored Wikipedia text embeddings, significantly reducing inference time per sample by a factor of 2,746 to 16,746. During the fine-tuning step, the time required for each sample is reduced by a factor of 23 to 67. Overall, the total training time shows a maximum reduction of 27.5 times across different datasets. Most importantly, our model has achieved SOTA (state-of-the-art) accuracy on two of the three commonly used datasets for evaluating zero-shot classification, namely the AG News (0.84) and Yahoo! Answers (0.64) datasets. The code for WC-SBERT is publicly available on GitHub 1 , and the dataset can also be accessed on Hugging Face 2 .
{"title":"WC-SBERT: Zero-Shot Topic Classification Using SBERT and Light Self-Training on Wikipedia Categories","authors":"Te-Yu Chi, Jyh-Shing Roger Jang","doi":"10.1145/3678183","DOIUrl":"https://doi.org/10.1145/3678183","url":null,"abstract":"\u0000 In NLP (natural language processing), zero-shot topic classification requires machines to understand the contextual meanings of texts in a downstream task without using the corresponding labeled texts for training, which is highly desirable for various applications [2]. In this paper, we propose a novel approach to construct a zero-shot task-specific model called WC-SBERT with satisfactory performance. The proposed approach is highly efficient since it uses light self-training requiring target labels (target class names of downstream tasks) only, which is distinct from other research that uses both the target labels and the unlabeled texts for training. In particular, during the pre-training stage, WC-SBERT uses contrastive learning with the multiple negative ranking loss [9] to construct the pre-trained model based on the similarity between Wiki categories. For the self-training stage, online contrastive loss is utilized to reduce the distance between a target label and Wiki categories of similar Wiki pages to the label. Experimental results indicate that compared to existing self-training models, WC-SBERT achieves rapid inference on approximately 6.45 million Wiki text entries by utilizing pre-stored Wikipedia text embeddings, significantly reducing inference time per sample by a factor of 2,746 to 16,746. During the fine-tuning step, the time required for each sample is reduced by a factor of 23 to 67. Overall, the total training time shows a maximum reduction of 27.5 times across different datasets. Most importantly, our model has achieved SOTA (state-of-the-art) accuracy on two of the three commonly used datasets for evaluating zero-shot classification, namely the AG News (0.84) and Yahoo! Answers (0.64) datasets. The code for WC-SBERT is publicly available on GitHub\u0000 \u0000 1\u0000 \u0000 , and the dataset can also be accessed on Hugging Face\u0000 \u0000 2\u0000 \u0000 .\u0000","PeriodicalId":48967,"journal":{"name":"ACM Transactions on Intelligent Systems and Technology","volume":null,"pages":null},"PeriodicalIF":7.2,"publicationDate":"2024-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141825634","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Moreno La Quatra, Giuseppe Gallipoli, Luca Cagliero
Text Style Transfer (TST) is a relevant branch of natural language processing that aims to control the style attributes of a piece of text while preserving its original content. To address TST in the absence of parallel data, Cycle-consistent Generative Adversarial Networks (CycleGANs) have recently emerged as promising solutions. Existing CycleGAN-based TST approaches suffer from the following limitations: (1) They apply self-supervision, based on the cycle-consistency principle, in the latent space. This approach turns out to be less robust to mixed-style inputs, i.e., when the source text is partly in the original and partly in the target style; (2) Generators and discriminators rely on recurrent networks, which are exposed to known issues with long-term text dependencies; (3) The target style is weakly enforced, as the discriminator distinguishes real from fake sentences without explicitly accounting for the generated text’s style. We propose a new CycleGAN-based TST approach that applies self-supervision directly at the sequence level to effectively handle mixed-style inputs and employs Transformers to leverage the attention mechanism for both text encoding and decoding. We also employ a pre-trained style classifier to guide the generation of text in the target style while maintaining the original content’s meaning. The experimental results achieved on the formality and sentiment transfer tasks show that our approach outperforms existing ones, both CycleGAN-based and not (including an open-source Large Language Model), on benchmark data and shows better robustness to mixed-style inputs.
{"title":"Self-supervised Text Style Transfer using Cycle-Consistent Adversarial Networks","authors":"Moreno La Quatra, Giuseppe Gallipoli, Luca Cagliero","doi":"10.1145/3678179","DOIUrl":"https://doi.org/10.1145/3678179","url":null,"abstract":"Text Style Transfer (TST) is a relevant branch of natural language processing that aims to control the style attributes of a piece of text while preserving its original content. To address TST in the absence of parallel data, Cycle-consistent Generative Adversarial Networks (CycleGANs) have recently emerged as promising solutions. Existing CycleGAN-based TST approaches suffer from the following limitations: (1) They apply self-supervision, based on the cycle-consistency principle, in the latent space. This approach turns out to be less robust to mixed-style inputs, i.e., when the source text is partly in the original and partly in the target style; (2) Generators and discriminators rely on recurrent networks, which are exposed to known issues with long-term text dependencies; (3) The target style is weakly enforced, as the discriminator distinguishes real from fake sentences without explicitly accounting for the generated text’s style. We propose a new CycleGAN-based TST approach that applies self-supervision directly at the sequence level to effectively handle mixed-style inputs and employs Transformers to leverage the attention mechanism for both text encoding and decoding. We also employ a pre-trained style classifier to guide the generation of text in the target style while maintaining the original content’s meaning. The experimental results achieved on the formality and sentiment transfer tasks show that our approach outperforms existing ones, both CycleGAN-based and not (including an open-source Large Language Model), on benchmark data and shows better robustness to mixed-style inputs.","PeriodicalId":48967,"journal":{"name":"ACM Transactions on Intelligent Systems and Technology","volume":null,"pages":null},"PeriodicalIF":7.2,"publicationDate":"2024-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141827121","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Meriem Arbaoui, Mohamed-el-Amine Brahmia, Abdellatif Rahmoun, M. Zghal
The emerging integration of IoT (Internet of Things) and AI (Artificial Intelligence) has unlocked numerous opportunities for innovation across diverse industries. However, growing privacy concerns and data isolation issues have inhibited this promising advancement. Unfortunately, traditional centralized machine learning (ML) methods have demonstrated their limitations in addressing these hurdles. In response to this ever-evolving landscape, Federated Learning (FL) has surfaced as a cutting-edge machine learning paradigm, enabling collaborative training across decentralized devices. FL allows users to jointly construct AI models without sharing their local raw data, ensuring data privacy, network scalability, and minimal data transfer. One essential aspect of FL revolves around proficient knowledge aggregation within a heterogeneous environment. Yet, the inherent characteristics of FL have amplified the complexity of its practical implementation compared to centralized ML. This survey delves into three prominent clusters of FL research contributions: personalization, optimization, and robustness. The objective is to provide a well-structured and fine-grained classification scheme related to these research areas through a unique methodology for selecting related work. Unlike other survey papers, we employed a hybrid approach that amalgamates bibliometric analysis and systematic scrutinizing to find the most influential work in the literature. Therefore, we examine challenges and contemporary techniques related to heterogeneity, efficiency, security, and privacy. Another valuable asset of this study is its comprehensive coverage of FL aggregation strategies, encompassing architectural features, synchronization methods, and several federation motivations. To further enrich our investigation, we provide practical insights into evaluating novel FL proposals and conduct experiments to assess and compare aggregation methods under IID and non-IID data distributions. Finally, we present a compelling set of research avenues that call for further exploration to open up a treasure of advancement.
{"title":"Federated Learning Survey: A Multi-Level Taxonomy of Aggregation Techniques, Experimental Insights, and Future Frontiers","authors":"Meriem Arbaoui, Mohamed-el-Amine Brahmia, Abdellatif Rahmoun, M. Zghal","doi":"10.1145/3678182","DOIUrl":"https://doi.org/10.1145/3678182","url":null,"abstract":"The emerging integration of IoT (Internet of Things) and AI (Artificial Intelligence) has unlocked numerous opportunities for innovation across diverse industries. However, growing privacy concerns and data isolation issues have inhibited this promising advancement. Unfortunately, traditional centralized machine learning (ML) methods have demonstrated their limitations in addressing these hurdles. In response to this ever-evolving landscape, Federated Learning (FL) has surfaced as a cutting-edge machine learning paradigm, enabling collaborative training across decentralized devices. FL allows users to jointly construct AI models without sharing their local raw data, ensuring data privacy, network scalability, and minimal data transfer. One essential aspect of FL revolves around proficient knowledge aggregation within a heterogeneous environment. Yet, the inherent characteristics of FL have amplified the complexity of its practical implementation compared to centralized ML. This survey delves into three prominent clusters of FL research contributions: personalization, optimization, and robustness. The objective is to provide a well-structured and fine-grained classification scheme related to these research areas through a unique methodology for selecting related work. Unlike other survey papers, we employed a hybrid approach that amalgamates bibliometric analysis and systematic scrutinizing to find the most influential work in the literature. Therefore, we examine challenges and contemporary techniques related to heterogeneity, efficiency, security, and privacy. Another valuable asset of this study is its comprehensive coverage of FL aggregation strategies, encompassing architectural features, synchronization methods, and several federation motivations. To further enrich our investigation, we provide practical insights into evaluating novel FL proposals and conduct experiments to assess and compare aggregation methods under IID and non-IID data distributions. Finally, we present a compelling set of research avenues that call for further exploration to open up a treasure of advancement.","PeriodicalId":48967,"journal":{"name":"ACM Transactions on Intelligent Systems and Technology","volume":null,"pages":null},"PeriodicalIF":7.2,"publicationDate":"2024-07-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141828352","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Lo Pang-Yun Ting, Huan-Yang Wang, Jhe-Yun Jhang, Kun-Ta Chuang
The growing adoption of electric vehicles (EVs) has resulted in an increased demand for public EV charging infrastructure. Currently, the collaboration between these stations has become vital for efficient charging scheduling and cost reduction. However, most existing scheduling methods primarily focus on recommending charging stations without considering users’ charging preferences. Adopting these strategies may require considerable modifications to how people charge their EVs, which could lead to a reluctance to follow the scheduling plan from charging services in real-world situations. To address these challenges, we propose the POSKID framework in this paper. It focuses on spatial-temporal charging scheduling, aiming to recommend a feasible charging arrangement, including a charging station and a charging time slot, to each EV user while minimizing overall operating costs and ensuring users’ charging satisfaction. The framework adopts an online charging mechanism that provides recommendations without prior knowledge of future electricity information or charging requests. To enhance users’ willingness to accept the recommendations, POSKID incorporates an incentive strategy and a novel embedding method combined with Bayesian personalized analysis. These techniques reveal users’ implicit charging preferences, enhancing the success probability of the charging scheduling task. Furthermore, POSKID integrates an online candidate arrangement selection and an explore-exploit strategy to improve the charging arrangement recommendations based on users’ feedback. Experimental results using real-world datasets validate the effectiveness of POSKID in optimizing charging management, surpassing other strategies. The results demonstrate that POSKID benefits each charging station while ensuring user charging satisfaction.
{"title":"Online Spatial-Temporal EV Charging Scheduling with Incentive Promotion","authors":"Lo Pang-Yun Ting, Huan-Yang Wang, Jhe-Yun Jhang, Kun-Ta Chuang","doi":"10.1145/3678180","DOIUrl":"https://doi.org/10.1145/3678180","url":null,"abstract":"\u0000 The growing adoption of electric vehicles (EVs) has resulted in an increased demand for public EV charging infrastructure. Currently, the collaboration between these stations has become vital for efficient charging scheduling and cost reduction. However, most existing scheduling methods primarily focus on recommending charging stations without considering users’ charging preferences. Adopting these strategies may require considerable modifications to how people charge their EVs, which could lead to a reluctance to follow the scheduling plan from charging services in real-world situations. To address these challenges, we propose the\u0000 POSKID\u0000 framework in this paper. It focuses on spatial-temporal charging scheduling, aiming to recommend a feasible charging arrangement, including a charging station and a charging time slot, to each EV user while minimizing overall operating costs and ensuring users’ charging satisfaction. The framework adopts an online charging mechanism that provides recommendations without prior knowledge of future electricity information or charging requests. To enhance users’ willingness to accept the recommendations,\u0000 POSKID\u0000 incorporates an incentive strategy and a novel embedding method combined with Bayesian personalized analysis. These techniques reveal users’ implicit charging preferences, enhancing the success probability of the charging scheduling task. Furthermore,\u0000 POSKID\u0000 integrates an online candidate arrangement selection and an explore-exploit strategy to improve the charging arrangement recommendations based on users’ feedback. Experimental results using real-world datasets validate the effectiveness of\u0000 POSKID\u0000 in optimizing charging management, surpassing other strategies. The results demonstrate that\u0000 POSKID\u0000 benefits each charging station while ensuring user charging satisfaction.\u0000","PeriodicalId":48967,"journal":{"name":"ACM Transactions on Intelligent Systems and Technology","volume":null,"pages":null},"PeriodicalIF":7.2,"publicationDate":"2024-07-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141828891","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Zhuang Yan, Junyan Zhang, Ruogu Lu, Kunlun He, Xiuxing Li
Ever-growing electronic medical corpora provide unprecedented opportunities for researchers to analyze patient conditions and drug effects. Meanwhile, severe challenges emerged in the large-scale electronic medical records process phase. Primarily, emerging words for medical terms, including informal descriptions, are difficult to recognize. Moreover, although deep models can help in entity extraction on medical texts, it requires large-scale labels which are time-intensive to obtain and not always available in the medical domain. However, when encountering a situation where massive unseen concepts appear, or labeled data is insufficient, the performance of existing algorithms will suffer an intolerable decline. In this paper, we propose a balanced and deep active learning framework ( MedNER ) for Named Entity Recognition in the medical corpus to alleviate above problems. Specifically, to describe our selection strategy precisely, we first define the uncertainty of a medical sentence as a labeling loss predicted by a loss-prediction module and define diversity as the least text distance between pairs of sentences in a sample batch computed based on word-morpheme embeddings. Furthermore, aiming to make a trade-off between uncertainty and diversity, we formulate a Distinct-K optimization problem to maximize the slightest uncertainty and diversity of chosen sentences. Finally, we propose a threshold-based approximation selection algorithm, Distinct-K Filter , which selects the most beneficial training samples by balancing diversity and uncertainty. Extensive experimental results on real datasets demonstrate that MedNER significantly outperforms existing approaches.
{"title":"MedNER: Enhanced Named Entity Recognition in Medical Corpus via Optimized Balanced and Deep Active Learning","authors":"Zhuang Yan, Junyan Zhang, Ruogu Lu, Kunlun He, Xiuxing Li","doi":"10.1145/3678178","DOIUrl":"https://doi.org/10.1145/3678178","url":null,"abstract":"\u0000 Ever-growing electronic medical corpora provide unprecedented opportunities for researchers to analyze patient conditions and drug effects. Meanwhile, severe challenges emerged in the large-scale electronic medical records process phase. Primarily, emerging words for medical terms, including informal descriptions, are difficult to recognize. Moreover, although deep models can help in entity extraction on medical texts, it requires large-scale labels which are time-intensive to obtain and not always available in the medical domain. However, when encountering a situation where massive unseen concepts appear, or labeled data is insufficient, the performance of existing algorithms will suffer an intolerable decline. In this paper, we propose a balanced and deep active learning framework (\u0000 MedNER\u0000 ) for Named Entity Recognition in the medical corpus to alleviate above problems. Specifically, to describe our selection strategy precisely, we first define the uncertainty of a medical sentence as a labeling loss predicted by a loss-prediction module and define diversity as the least text distance between pairs of sentences in a sample batch computed based on word-morpheme embeddings. Furthermore, aiming to make a trade-off between uncertainty and diversity, we formulate a\u0000 Distinct-K\u0000 optimization problem to maximize the slightest uncertainty and diversity of chosen sentences. Finally, we propose a threshold-based approximation selection algorithm,\u0000 Distinct-K Filter\u0000 , which selects the most beneficial training samples by balancing diversity and uncertainty. Extensive experimental results on real datasets demonstrate that\u0000 MedNER\u0000 significantly outperforms existing approaches.\u0000","PeriodicalId":48967,"journal":{"name":"ACM Transactions on Intelligent Systems and Technology","volume":null,"pages":null},"PeriodicalIF":7.2,"publicationDate":"2024-07-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141831230","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
As users’ social media feeds have become increasingly driven by algorithmically recommended content, there is a need to understand the impact these recommendations have on users. People in recovery from eating disorders (ED) may try to avoid content that features severely underweight bodies or that encourages disordered eating. However, if recommender systems show them this type of content anyway, it may impact their recovery or even lead to relapse. In this study, we take a two-pronged approach to understanding the intersection of recommender systems, eating disorder content, and users in recovery. We performed a content analysis of tweets about recommended eating disorder content and conducted a small-scale study on Pinterest to show that eating disorder content is recommended in response to interaction with posts about eating disorder recovery. We discuss the implications for responsible recommendation and harm prevention.
{"title":"Recommender System-Induced Eating Disorder Relapse: Harmful Content and the Challenges of Responsible Recommendation","authors":"Jennifer Golbeck","doi":"10.1145/3675404","DOIUrl":"https://doi.org/10.1145/3675404","url":null,"abstract":"As users’ social media feeds have become increasingly driven by algorithmically recommended content, there is a need to understand the impact these recommendations have on users. People in recovery from eating disorders (ED) may try to avoid content that features severely underweight bodies or that encourages disordered eating. However, if recommender systems show them this type of content anyway, it may impact their recovery or even lead to relapse. In this study, we take a two-pronged approach to understanding the intersection of recommender systems, eating disorder content, and users in recovery. We performed a content analysis of tweets about recommended eating disorder content and conducted a small-scale study on Pinterest to show that eating disorder content is recommended in response to interaction with posts about eating disorder recovery. We discuss the implications for responsible recommendation and harm prevention.","PeriodicalId":48967,"journal":{"name":"ACM Transactions on Intelligent Systems and Technology","volume":null,"pages":null},"PeriodicalIF":7.2,"publicationDate":"2024-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141673165","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Saleh Baghersalimi, A. Amirshahi, Farnaz Forooghifar, T. Teijeiro, A. Aminifar, David Atienza
Integrating low-power wearable systems into routine health monitoring is an ongoing challenge. Recent advances in the computation capabilities of wearables make it possible to target complex scenarios by exploiting multiple biosignals and using high-performance algorithms, such as Deep Neural Networks (DNNs). However, there is a trade-off between the algorithms’ performance and the low-power requirements of platforms with limited resources. Besides, physically larger and multi-biosignal-based wearables bring significant discomfort to the patients. Consequently, reducing power consumption and discomfort is necessary for patients to use wearable devices continuously during everyday life. To overcome these challenges, in the context of epileptic seizure detection, we propose the M2SKD (Multi-to-Single Knowledge Distillation) approach targeting single-biosignal processing in wearable systems. The starting point is to train a highly-accurate multi-biosignal DNN, then apply M2SKD to develop a single-biosignal DNN solution for wearable systems that achieves an accuracy comparable to the original multi-biosignal DNN. To assess the practicality of our approach to real-life scenarios, we perform a comprehensive simulation experiment analysis on several edge computing platforms.
{"title":"M2SKD: Multi-to-Single Knowledge Distillation of Real-Time Epileptic Seizure Detection for Low-Power Wearable Systems","authors":"Saleh Baghersalimi, A. Amirshahi, Farnaz Forooghifar, T. Teijeiro, A. Aminifar, David Atienza","doi":"10.1145/3675402","DOIUrl":"https://doi.org/10.1145/3675402","url":null,"abstract":"Integrating low-power wearable systems into routine health monitoring is an ongoing challenge. Recent advances in the computation capabilities of wearables make it possible to target complex scenarios by exploiting multiple biosignals and using high-performance algorithms, such as Deep Neural Networks (DNNs). However, there is a trade-off between the algorithms’ performance and the low-power requirements of platforms with limited resources. Besides, physically larger and multi-biosignal-based wearables bring significant discomfort to the patients. Consequently, reducing power consumption and discomfort is necessary for patients to use wearable devices continuously during everyday life. To overcome these challenges, in the context of epileptic seizure detection, we propose the M2SKD (Multi-to-Single Knowledge Distillation) approach targeting single-biosignal processing in wearable systems. The starting point is to train a highly-accurate multi-biosignal DNN, then apply M2SKD to develop a single-biosignal DNN solution for wearable systems that achieves an accuracy comparable to the original multi-biosignal DNN. To assess the practicality of our approach to real-life scenarios, we perform a comprehensive simulation experiment analysis on several edge computing platforms.","PeriodicalId":48967,"journal":{"name":"ACM Transactions on Intelligent Systems and Technology","volume":null,"pages":null},"PeriodicalIF":7.2,"publicationDate":"2024-07-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141677621","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Explainable recommender systems (ERS) aim to enhance users’ trust in the systems by offering personalized recommendations with transparent explanations. This transparency provides users with a clear understanding of the rationale behind the recommendations, fostering a sense of confidence and reliability in the system’s outputs. Generally, the explanations are presented in a familiar and intuitive way, which is in the form of natural language, thus enhancing their accessibility to users. Recently, there has been an increasing focus on leveraging reviews as a valuable source of rich information in both modeling user-item preferences and generating textual interpretations, which can be performed simultaneously in a multi-task framework. Despite the progress made in these review-based recommendation systems, the integration of implicit feedback derived from user-item interactions and user-written text reviews has yet to be fully explored. To fill this gap, we propose a model named SERMON (Aspect-enhanced Explainable Recommendation with Multi-modal Contrast Learning). Our model explores the application of multimodal contrastive learning to facilitate reciprocal learning across two modalities, thereby enhancing the modeling of user preferences. Moreover, our model incorporates the aspect information extracted from the review, which provides two significant enhancements to our tasks. Firstly, the quality of the generated explanations is improved by incorporating the aspect characteristics into the explanations generated by a pre-trained model with controlled textual generation ability. Secondly, the commonly used user-item interactions are transformed into user-item-aspect interactions, which we refer to as interaction triple, resulting in a more nuanced representation of user preference. To validate the effectiveness of our model, we conduct extensive experiments on three real-world datasets. The experimental results show that our model outperforms state-of-the-art baselines, with a 2.0% improvement in prediction accuracy and a substantial 24.5% enhancement in explanation quality for the TripAdvisor dataset.
{"title":"Aspect-enhanced Explainable Recommendation with Multi-modal Contrastive Learning","authors":"Hao Liao, Shuo Wang, Hao Cheng, Wei Zhang, Jiwei Zhang, Mingyang Zhou, Kezhong Lu, Rui Mao, Xing Xie","doi":"10.1145/3673234","DOIUrl":"https://doi.org/10.1145/3673234","url":null,"abstract":"<p>Explainable recommender systems (<b>ERS</b>) aim to enhance users’ trust in the systems by offering personalized recommendations with transparent explanations. This transparency provides users with a clear understanding of the rationale behind the recommendations, fostering a sense of confidence and reliability in the system’s outputs. Generally, the explanations are presented in a familiar and intuitive way, which is in the form of natural language, thus enhancing their accessibility to users. Recently, there has been an increasing focus on leveraging reviews as a valuable source of rich information in both modeling user-item preferences and generating textual interpretations, which can be performed simultaneously in a multi-task framework. Despite the progress made in these review-based recommendation systems, the integration of implicit feedback derived from user-item interactions and user-written text reviews has yet to be fully explored. To fill this gap, we propose a model named <b>SERMON</b> (A<b><underline>s</underline></b>pect-enhanced <b><underline>E</underline></b>xplainable <b><underline>R</underline></b>ecommendation with <b><underline>M</underline></b>ulti-modal C<b><underline>o</underline></b>ntrast Lear<b><underline>n</underline></b>ing). Our model explores the application of multimodal contrastive learning to facilitate reciprocal learning across two modalities, thereby enhancing the modeling of user preferences. Moreover, our model incorporates the aspect information extracted from the review, which provides two significant enhancements to our tasks. Firstly, the quality of the generated explanations is improved by incorporating the aspect characteristics into the explanations generated by a pre-trained model with controlled textual generation ability. Secondly, the commonly used user-item interactions are transformed into user-item-aspect interactions, which we refer to as interaction triple, resulting in a more nuanced representation of user preference. To validate the effectiveness of our model, we conduct extensive experiments on three real-world datasets. The experimental results show that our model outperforms state-of-the-art baselines, with a 2.0% improvement in prediction accuracy and a substantial 24.5% enhancement in explanation quality for the TripAdvisor dataset.</p>","PeriodicalId":48967,"journal":{"name":"ACM Transactions on Intelligent Systems and Technology","volume":null,"pages":null},"PeriodicalIF":5.0,"publicationDate":"2024-06-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141508268","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：