Graph unlearning technology has become increasingly important since the advent of the `right to be forgotten' and the growing concerns about the privacy and security of artificial intelligence. Graph unlearning aims to quickly eliminate the effects of specific data on graph neural networks (GNNs). However, most existing deterministic graph unlearning frameworks follow a balanced partition-submodel training-aggregation paradigm, resulting in a lack of structural information between subgraph neighborhoods and redundant unlearning parameter calculations. To address this issue, we propose a novel Graph Structure Mapping Unlearning paradigm (GSMU) and a novel method based on it named Community-centric Graph Eraser (CGE). CGE maps community subgraphs to nodes, thereby enabling the reconstruction of a node-level unlearning operation within a reduced mapped graph. CGE makes the exponential reduction of both the amount of training data and the number of unlearning parameters. Extensive experiments conducted on five real-world datasets and three widely used GNN backbones have verified the high performance and efficiency of our CGE method, highlighting its potential in the field of graph unlearning.
FiCABU: A Fisher-Based, Context-Adaptive Machine Unlearning Processor for Edge AI
School of Intelligent Semiconductor Engineering, Chung-Ang University（智能半导体工程学院， Chung-Ang 大学）；AI Edge SoC Research Section, the Electronics and Telecommunications Research Institute（AI 边缘 SoC 研究部门，电子电信研究所）
Michigan State University（密歇根州立大学）；IBM Research（IBM研究院）；Cisco Research（思科研发）；University of California, Santa Cruz（加州大学圣克鲁兹分校）；ByteDance Research（字节跳动研究）；Carnegie Mellon University（卡内基梅隆大学）
University of Illinois at Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）；IBM T.J. Watson Research Center（IBM托马斯·J·沃森研究中心）；NVIDIA（英伟达）
With the increasing adoption of graph neural networks (GNNs) in the machine learning community, GPUs have become an essential tool to accelerate GNN training. However, training GNNs on very large graphs that do not fit in GPU memory is still a challenging task. Unlike conventional neural networks, mini-batching input samples in GNNs requires complicated tasks such as traversing neighboring nodes and gathering their feature values. While this process accounts for a significant portion of the training time, we find existing GNN implementations using popular deep neural network (DNN) libraries su…
Faculty of Information Technology, University of Science, Ho Chi Minh City, Vietnam.（信息科技学院，科学大学，胡志明市，越南）；Vietnam National University, Ho Chi Minh City, Vietnam.（越南国家大学，胡志明市，越南）
Towards Federated Domain Unlearning: Verification Methodologies and Challenges
University of Macau（澳门大学）；University Of Electronic Science And Technology（电子科技大学）；Institute of High Performance Computing, Agency for Science, Technology and Research（新加坡科技研究局高性能计算研究院）
LMU Munich（慕尼黑大学）；Munich Center for Machine Learning (MCML)（慕尼黑机器学习中心）；Technical University of Munich（慕尼黑技术大学）；Siemens AG（西门子股份公司）；University of Oxford（牛津大学）
The community lacks theory-informed guidelines for building good data sets. We analyse theoretical directions relating to what aspects of the data matter and conclude that the intuitions derived from the existing literature are incorrect and misleading. Using empirical counter-examples, we show that 1) data dimension should not necessarily be minimised and 2) when manipulating data, preserving the distribution is inessential. This calls for a more data-aware theoretical understanding. Although not explored in this work, we propose the study of the impact of data modification on learned represe…
DataPerf: Benchmarks for Data-Centric AI Development
Harvard University（哈佛大学）；ETH Zurich（苏黎世联邦理工学院）；Landing AI（Landing AI公司）；Google（谷歌）；University of Oxford（牛津大学）；Carnegie Mellon University（卡内基梅隆大学）；Stanford University（斯坦福大学）；Contextual AI（Contextual AI公司）；MLCommons（MLCommons组织）；Factored（Factored公司）；Meta；Cohere（Cohere公司）；University College London（伦敦大学学院）；University of Zurich（苏黎世大学）；San Diego Supercomputer Center, UC San Diego（加州大学圣地亚哥分校圣地亚哥超级计算机中心）；Cleanlab（Cleanlab公司）；Eindhoven University of Technology（埃因霍温理工大学）；Institute for Human and Machine Cognition（人类与机器认知研究所）
Data-centric AI is a new and exciting research topic in the AI community, but many organizations already build and maintain various "data-centric" applications whose goal is to produce high quality data. These range from traditional business data processing applications (e.g., "how much should we charge each of our customers this month?") to production ML systems such as recommendation engines. The fields of data and ML engineering have arisen in recent years to manage these applications, and both include many interesting novel tools and processes. In this paper, we discuss several lessons fro…
Data-Centric AI in the Age of Large Language Models
National University of Singapore（新加坡国立大学）；Agency for Science, Technology and Research（新加坡科技研究局）；Singapore-MIT Alliance for Research and Technology Centre（新加坡-麻省理工研究与技术联盟中心）；Guangdong Lab of AI and Digital Economy (SZ)（广东人工智能与数字经济实验室（深圳））；Massachusetts Institute of Technology（麻省理工学院）；Allen Institute for AI（艾伦人工智能研究所）；University of Washington（华盛顿大学）
College of Computing, Georgia Institute of Technology（佐治亚理工学院计算学院）；Faculty of Science, University of Amsterdam（阿姆斯特丹大学理学院）；Department of Theatre, Northeastern University（东北大学戏剧系）；Lawrence Berkeley National Laboratory（劳伦斯伯克利国家实验室）
Data-centric AI is at the center of a fundamental shift in software engineering where machine learning becomes the new software, powered by big data and computing infrastructure. Here software engineering needs to be re-thought where data becomes a first-class citizen on par with code. One striking observation is that a significant portion of the machine learning process is spent on data preparation. Without good data, even the best machine learning algorithms cannot perform well. As a result, data-centric AI practices are now becoming mainstream. Unfortunately, many datasets in the real world…
With the progressive commoditization of modeling capabilities, data-centric AI recognizes that what happens before and after training becomes crucial for real-world deployments. Following the intuition behind Model Cards, we propose DAG Cards as a form of documentation encompassing the tenets of a data-centric point of view. We argue that Machine Learning pipelines (rather than models) are the most appropriate level of documentation for many practical use cases, and we share with the community an open implementation to generate cards from code.
Exabyte Inc.（Exabyte公司）；University of California, Berkeley（加州大学伯克利分校）
The advent of data-driven science in the 21st century brought about the need for well-organized structured data and associated infrastructure able to facilitate the applications of Artificial Intelligence and Machine Learning. We present an effort aimed at organizing the diverse landscape of physics-based and data-driven computational models in order to facilitate the storage of associated information as structured data. We apply object-oriented design concepts and outline the foundations of an open-source collaborative framework that is: (1) capable of uniquely describing the approaches in st…
Nanyang Technological University（南洋理工大学）；SenseTime Research（商汤科技研究院）；Shanghai AI Laboratory（上海人工智能实验室）
Unconditional human image generation is an important task in vision and graphics, which enables various applications in the creative industry. Existing studies in this field mainly focus on "network engineering" such as designing new components and objective functions. This work takes a data-centric perspective and investigates multiple critical aspects in "data engineering", which we believe would complement the current practice. To facilitate a comprehensive study, we collect and annotate a large-scale human image dataset with over 230K samples capturing diverse poses and textures. Equipped …
Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences（中国科学院深圳先进技术研究院）；University of Chinese Academy of Sciences（中国科学院大学）；University of Macau（澳门大学）
Unlearnable example attacks are data poisoning techniques that can be used to safeguard public data against unauthorized use for training deep learning models. These methods add stealthy perturbations to the original image, thereby making it difficult for deep learning models to learn from these training data effectively. Current research suggests that adversarial training can, to a certain degree, mitigate the impact of unlearnable example attacks, while common data augmentation methods are not effective against such poisons. Adversarial training, however, demands considerable computational r…
Towards deployment-centric multimodal AI beyond vision and language
Centre for Machine Intelligence, University of Sheffield, Sheffield, UK（机器智能中心，谢菲尔德大学）；School of Computer Science, University of Sheffield, Sheffield, UK（计算机科学学院，谢菲尔德大学）；The Alan Turing Institute, London, UK（艾伦·图灵研究所，伦敦）；Department of Metabolism, Digestion and Reproduction, Imperial College London, London, UK（代谢、消化与生殖部门，伦敦帝国学院）；Department of Applied AI, Simula Research Laboratory, Oslo, Norway（应用人工智能部门，Simula研究实验室，奥斯陆）；Information School, University of Sheffield, Sheffield, UK（信息学院，谢菲尔德大学）；NHS England, Leeds, UK（英国英格兰国家医疗服务体系，利兹）；Institute of Health Informatics, University College London, London, UK（健康信息研究所，伦敦大学学院）；Department of Engineering, King’s College London, London, UK（工程部门，伦敦国王学院）；Department of Computing Science, University of Aberdeen, Aberdeen, UK（计算科学部门，阿伯丁大学）；School of Informatics, University of Edinburgh, Edinburgh, UK（信息学院，爱丁堡大学）；School of Engineering Mathematics and Technology, University of Bristol, Bristol, UK（工程数学与技术学院，布里斯托尔大学）；Department of Informatics, King’s College London, London, UK（信息部门，伦敦国王学院）；Department of Earth Sciences, University of Cambridge, Cambridge, UK（地球科学部门，剑桥大学）；Department of Computer Science, University of Manchester, Manchester, UK（计算机科学部门，曼彻斯特大学）
DMLR: Data-centric Machine Learning Research -- Past, Present and Future
ByteDance Research（字节跳动研究院）；ICSI（国际计算机科学研究所）；LBNL（劳伦斯伯克利国家实验室）；Fraunhofer HHI（弗劳恩霍夫海因里希·赫兹研究所）；University of Chicago（芝加哥大学）；Microsoft AI for Good Lab（微软AI公益实验室）；University of Paris-Saclay（巴黎萨克雷大学）；MLCommons；Dotphoton；NASA（美国国家航空航天局）；Mod Op；Google（谷歌）；TU Delft（代尔夫特理工大学）；Carnegie Mellon University（卡内基梅隆大学）；UC Santa Cruz（加州大学圣克鲁兹分校）；University of Haifa（海法大学）；Fraunhofer IPA（弗劳恩霍夫制造工程与自动化研究所）；Cohere；CoactiveAI；Talon；UC Berkeley（加州大学伯克利分校）；Kaggle；UT Austin（德克萨斯大学奥斯汀分校）；TU Berlin（柏林工业大学）；Nutanix；Cleanlab；Snorkel AI；Harvard University（哈佛大学）；Princeton University（普林斯顿大学）；Stanford University（斯坦福大学）；University of Cambridge（剑桥大学）；Together；TU Eindhoven（埃因霍温理工大学）；ETH Zurich（苏黎世联邦理工学院）；ChaLearn
MERLIN: Multi-agent offline and transfer learning for occupant-centric energy flexible operation of grid-interactive communities using smart meter data and CityLearn
The University of Texas at Austin（德克萨斯大学奥斯汀分校）；Electric Power Research Institute（电力科学研究院）
