我们提出了一种新的和可扩展的标签嵌入框架,用于大规模的多标签学习a. k. aExMLDS(使用分布式语义的极端多标签学习)。我们的方法从分布式语义中汲取灵感,特别是跳过革兰氏阴性采样(SGNS)方法,广泛用于学习自然语言处理任务的词嵌入。学习这样的嵌入可以简化为某种矩阵分解。我们的方法是新颖的,因为它突出了用于多标签学习的标签嵌入方法和通常用于学习文本数据表示的段落/文档嵌入方法之间的有趣联系。该框架也可以很容易地扩展到包含辅助信息,如标签-标签相关性;这是至关重要的,特别是当训练数据中有很多缺失的标签时。我们通过在各种基准数据集上进行的大量实验证明了我们方法的有效性,并表明与几种基线和最先进的大规模多标签学习方法相比,所提出的学习方法表现良好。为了促进端到端的学习,我们开发了一个联合学习算法,可以学习嵌入以及回归模型,预测这些嵌入给定的输入功能,通过有效的基于梯度的方法。
英文摘要:
We present a novel and scalable label embedding framework for large-scale multi-label learning a.k.a ExMLDS (Extreme Multi-Label Learning using Distributional Semantics). Our approach draws inspiration from ideas rooted in distributional semantics, specifically the Skip Gram Negative Sampling (SGNS) approach, widely used to learn word embeddings for natural language processing tasks. Learning such embeddings can be reduced to a certain matrix factorization. Our approach is novel in that it highlights interesting connections between label embedding methods used for multi-label learning and paragraph/document embedding methods commonly used for learning representations of text data. The framework can also be easily extended to incorporate auxiliary information such as label-label correlations; this is crucial especially when there are a lot of missing labels in the training data. We demonstrate the effectiveness of our approach through an extensive set of experiments on a variety of benchmark datasets, and show that the proposed learning methods perform favorably compared to several baselines and state-of-the-art methods for large-scale multi-label learning. To facilitate end-to-end learning, we develop a joint learning algorithm that can learn the embeddings as well as a regression model that predicts these embeddings given input features, via efficient gradient-based methods.
The Dataset Nutrition Label (2nd Gen): Leveraging Context to Mitigate Harms in Artificial Intelligence
Berkman Klein Center(伯克曼·克莱因中心);Harvard University(哈佛大学);metaLAB;Data Nutrition Project(数据营养项目);MIT Quest for Intelligence(麻省理工学院智能探索计划);MIT(麻省理工学院);Massachusetts College of Liberal Arts(马萨诸塞文理学院)
Improving Multi-Label Contrastive Learning by Leveraging Label Distribution
College of Computer Science and Software Engineering, Hohai University(河海大学计算机与软件学院);State Key Laboratory for Novel Software Technology, Nanjing University(南京大学计算机软件新技术国家重点实验室)
Label-Assemble: Leveraging Multiple Datasets with Partial Labels
University of Illinois Urbana-Champaign(伊利诺伊大学厄巴纳-香槟分校);Johns Hopkins University(约翰斯·霍普金斯大学);Tongji University(同济大学);Johns Hopkins University School of Medicine(约翰斯·霍普金斯大学医学院)
Leveraging Label Potential for Enhanced Multimodal Emotion Recognition
Xinjiang University(新疆大学);School of Computer Science and Technology(计算机科学与技术学院);Xinjiang Multimodal Intelligent Processing and Information Security Engineering Technology Research Center(新疆多模态智能处理与信息安全工程技术研究中心)
BlinDNO: A Distributional Neural Operator for Dynamical System Reconstruction from Time-Label-Free data
Department of Mathematical Sciences, Tsinghua University, Beijing, China(清华大学数学科学系);Yau Mathematical Sciences Center, Tsinghua University, Beijing, China(清华大学姚期智信息科学研究院);Yanqi Lake Beijing Institute of Mathematical Sciences and Applications, Beijing, China(北京燕琦湖数学科学与应用研究院)
Leveraging Label Correlations in a Multi-label Setting: A Case Study in Emotion
University of Southern California(南加州大学);Information Science Institute, University of Southern California(南加州大学信息科学研究所);Microsoft Cognitive Services(微软认知服务)
检测文本中表达的情感已经成为一系列领域的关键。在这项工作中,我们研究如何利用标签相关性的多标签情感识别模型,以提高情感检测。首先,我们开发了两种建模方法来解决这个问题,以捕捉情感词本身的词关联,通过在输入中包含情感,或者利用Masked Language Modeling(MLM)。其次,我们将情感表示的成对约束作为正则化项与模型的分类损失相结合。我们将这些术语分为两类,局部和全局。前者基于黄金标签动态变化,而后者在训练期间保持静态。我们使用基于单语BERT的模型在SemEval 2018任务1 E-c中展示了西班牙语,英语和阿拉伯语的最先进性能。在更好的性能之上,我们还展示了更好的鲁棒性。代码可在https://github.com/gchochla/Demux-MEmo上获得。
Achieving Predictive Precision: Leveraging LSTM and Pseudo Labeling for Volvo's Discovery Challenge at ECML-PKDD 2024
ISTI-CNR;IMT School for Advanced Studies Lucca(卢cca高级研究学院);KU Leuven(鲁汶大学);Chalmers University of Technology(查尔姆斯理工大学);University of Pisa(比萨大学);University of Parma(帕尔马大学);University of Chieti-Pescara(基耶蒂-佩斯卡拉大学)
Leveraging weak complementary labels to improve semantic segmentation of hepatocellular carcinoma and cholangiocarcinoma in H&E-stained slides
Technische Universität Berlin(柏林工业大学);Aignostics GmbH(Aignostics 公司);Charité Universitätsmedizin Berlin(柏林夏里特医学院);Berlin Institute of Health at Charité – Universitätsmedizin Berlin(柏林健康研究所(夏里特医学院));University Hospital Düsseldorf(杜塞尔多夫大学医院);Ludwig-Maximilians-Universität München(慕尼黑大学);German Cancer Research Center(德国癌症研究中心)