This is an autogenerated issue. It uses the facebook/bart-large-cnn model to generate summaries for each papers appeared below.
Keyword: IoT (6 papers)
Spatiotemporal-Augmented Graph Neural Networks for Human Mobility Simulation
:brain: Authors: Yu Wang, Tongya Zheng, Shunyu Liu, Kaixuan Chen, Zunlei Feng, Yunzhi Hao, Mingli Song :paw_prints: Details:arXiv:2306.09381 :ramen: tl;dr: Human mobility patterns have shown significant applications in policy-decision scenarios and economic behavior researches. Existing methods mostly rely on the static relationships of locations. We propose a novel framework to model the dynamic spatiotemporal effects of locations, namely SpatioTemporal-Augmented gRaph neural networks (STAR)... :seedling: Read more →
Adaptive Hierarchical SpatioTemporal Network for Traffic Forecasting
:brain: Authors: Yirong Chen, Ziyue Li, Wanli Ouyang, Michael Lepech :paw_prints: Details:arXiv:2306.09386 :ramen: tl;dr: Accurate traffic forecasting is vital to intelligent transportation systems, which are widely adopted to solve urban traffic issues. Existing traffic forecasting studies focus on modeling spatial-temporal dynamics in traffic data. In this work, we propose an Adaptive Hierarchical SpatioTemporal Network (AHSTN) to promote traffic forecasting.... :sunflower: Comments: Accepted to IEEE CASE 2023 :seedling: Read more →
Multi-Objective and Model-Predictive Tree Search for Spatiotemporal Informative Planning
:brain: Authors: Weizhe Chen, Lantao Liu :paw_prints: Details:arXiv:2306.09608 :ramen: tl;dr: Adaptive sampling and planning in robotic environmental monitoring are challenging when the target environmental process varies over space and time. We propose a Monte Carlo tree search method which not only well balances the environment exploration and exploitation in space, but also catches up to the temporal environmental dynamics. This is achieved by incorporating multi-objective optimization and a look-ahead model-predictive rewarding mechanism.... :sunflower: Comments: Accepted to the 58th IEEE Conference on Decision and Control (CDC 2019) :seedling: Read more →
EVOPOSE: A Recursive Transformer For 3D Human Pose Estimation With Kinematic Structure Priors
:brain: Authors: Yaqi Zhang, Yan Lu, Bin Liu, Zhiwei Zhao, Qi Chu, Nenghai Yu :paw_prints: Details:arXiv:2306.09615 :ramen: tl;dr: Transformer is popular in recent 3D human pose estimation, which utilizes long-term modeling to lift 2D keypoints into the 3D space. Current transformer-based methods do not fully exploit the prior knowledge of the human skeleton provided by the kinematic structure. In this paper, we propose a novel transformer- based model EvoPose to introduce the human body prior knowledge.... :sunflower: Comments: 5 pages, 2 figures, 4 tables, published in the proceedings of IEEE ICASSP 2023 :seedling: Read more →
Dynamic Decision Tree Ensembles for Energy-Efficient Inference on IoT Edge Nodes
:brain: Authors: Francesco Daghero, Alessio Burrello, Enrico Macii, Paolo Montuschi, Massimo Poncino, Daniele Jahier Pagliari :paw_prints: Details:arXiv:2306.09789 :ramen: tl;dr: With the increasing popularity of Internet of Things (IoT) devices, there is a growing need for energy-efficient Machine Learning (ML) models that can run on constrained edge nodes. Decision tree ensembles, such as Random Forests (RFs) and Gradient Boosting (GBTs), are particularly suited for this task.... :sunflower: Comments: This article has been accepted for publication in IEEE Internet of Things Journal :seedling: Read more →
PanoOcc: Unified Occupancy Representation for Camera-based 3D Panoptic Segmentation
:brain: Authors: Yuqi Wang, Yuntao Chen, Xingyu Liao, Lue Fan, Zhaoxiang Zhang :paw_prints: Details:arXiv:2306.10013 :ramen: tl;dr: PanoOcc uses voxel queries to aggregate spatiotemporal information from multi-frame and multi-view images in a coarse-to-fine scheme. It integrates feature learning and scene representation into a unified occupancy representation. The code will be released at https //github.com/Robertwyq/Pano Occ.... :sunflower: Comments: technical report :seedling: Read more →
Keyword: edge computing (3 papers)
A flexible algorithm to offload DAG applications for edge computing
:brain: Authors: Gabriel F. C. de Queiroz, José F. de Rezende, Valmir C. Barbosa :paw_prints: Details:arXiv:2306.09458 :ramen: tl;dr: Multi-access Edge Computing (MEC) is an enabling technology to leverage new network applications, such as virtual/augmented reality. MEC provides faster task processing at the network edge. This is done by deploying servers closer to the end users to run the network applications.... :seedling: Read more →
A Design Approach and Prototype Implementation for Factory Monitoring Based on Virtual and Augmented Reality at the Edge of Industry 4.0
:brain: Authors: Christos Anagnostopoulos, Georgios Mylonas, Apostolos P. Fournaris, Christos Koulamas :paw_prints: Details:arXiv:2306.09692 :ramen: tl;dr: We present our approach for a monitoring intervention inside a factory space utilizing both VR and AR. We discuss its main design directions, as well as a basic ontology to aid in simple description of factory assets. In order to highlight the design aspects of our approach, we present a prototype implementation, based on a use case scenario in a factory site.... :sunflower: Comments: Preprint submitted to the 21st IEEE International Conference on Industrial Informatics, INDIN'23 :seedling: Read more →
Can Orbital Servers Provide Mars-Wide Edge Computing?
:brain: Authors: Tobias Pfandzelter, David Bermbach :paw_prints: Details:arXiv:2306.09756 :ramen: tl;dr: Human landing, exploration and settlement on Mars will require local compute resources at the Mars edge. Landing such resources on Mars is an expensive endeavor. Instead, concepts from low-Earth orbit edge computing may be applied to Mars edge computing. This could lower launching costs of compute resources for Mars while also providing Mars-wide networking.... :seedling: Read more →
Keyword: Blockchain (2 papers)
PIEChain -- A Practical Blockchain Interoperability Framework
:brain: Authors: Daniël Reijsbergen, Aung Maw, Jingchi Zhang, Tien Tuan Anh Dinh, Anwitaman Datta :paw_prints: Details:arXiv:2306.09735 :ramen: tl;dr: PIEChain is a general, Kafka-based cross-chain communication framework. We utilize PIEChain for a practical case study a cross- Chain auction. Users who hold tokens on multiple chains bid for a ticket sold on another chain. PIE chain is the first publicly available, practical implementation of a general framework for cross-Chain communication.... :seedling: Read more →
CroCoDai: A Stablecoin for Cross-Chain Commerce
:brain: Authors: Daniël Reijsbergen, Bretislav Hajek, Tien Tuan Anh Dinh, Jussi Keppo, Hank Korth, Anwitaman Datta :paw_prints: Details:arXiv:2306.09754 :ramen: tl;dr: Decentralized Finance (DeFi) has grown rapidly in value in recent years. The global DeFi ecosystem is fragmented into multiple blockchains, fueling the demand for cross-chain commerce. Stablecoins, which are pegged to a non-volatile asset such as the US dollar, help mitigate the risk associated with price fluctuations.... :seedling: Read more →
Keyword: Federated Learning (7 papers)
A Simple Data Augmentation for Feature Distribution Skewed Federated Learning
:brain: Authors: Yunlu Yan, Lei Zhu :paw_prints: Details:arXiv:2306.09363 :ramen: tl;dr: Federated learning (FL) facilitates collaborative learning among multiple clients. Main challenge lies in the feature shift caused by the different underlying distributions of local datasets. FedRDN is a plug-and-play component, which can be seamlessly integrated into the data augmentation flow with only a few lines of code.... :sunflower: Comments: 11 pages, 3 figures :seedling: Read more →
Towards Practical Federated Causal Structure Learning
:brain: Authors: Zhaoyu Wang, Pingchuan Ma, Shuai Wang :paw_prints: Details:arXiv:2306.09433 :ramen: tl;dr: The process of causal structure learning involves identifying causal graphs from observational data. FedC2SL is a federated constraint-based causal structurelearning scheme that learns causal graphs. It examines conditional independence between two variables under a condition set without collecting raw data from clients. The study finds it demonstrates encouraging performance and strong resilience to data heterogeneity among clients.... :seedling: Read more →
Opportunistic Transmission of Distributed Learning Models in Mobile UAVs
:brain: Authors: Jingxin Li, Xiaolan Liu, Toktam Mahmoodi :paw_prints: Details:arXiv:2306.09484 :ramen: tl;dr: Proposal aims to take advantage of the proximity of users to the base station or the general condition of the wireless transmission channel. During the training, intermediate model parameters are uploaded to the server, opportunistically and based on the wireless channel condition. Test accuracy increases by up to 13.47% with just one round of extra transmission.... :seedling: Read more →
FedMultimodal: A Benchmark For Multimodal Federated Learning
:brain: Authors: Tiantian Feng, Digbalay Bose, Tuo Zhang, Rajat Hebbar, Anil Ramakrishna, Rahul Gupta, Mi Zhang, Salman Avestimehr, Shrikanth Narayanan :paw_prints: Details:arXiv:2306.09486 :ramen: tl;dr: Federated Learning (FL) has become an emerging machine learning technique to tackle data privacy challenges through collaborative training. Despite significant efforts that have been made to FL in fields like computer vision, audio, and natural language processing, the FL applications utilizing multimodal data streams remain largely unexplored.... :sunflower: Comments: This paper was accepted to KDD 2023 Applied Data Science (ADS) track :seedling: Read more →
Fedstellar: A Platform for Decentralized Federated Learning
:brain: Authors: Enrique Tomás Martínez Beltrán, Ángel Luis Perales Gómez, Chao Feng, Pedro Miguel Sánchez Sánchez, Sergio López Bernal, Gérôme Bovet, Manuel Gil Pérez, Gregorio Martínez Pérez, Alberto Huertas Celdrán :paw_prints: Details:arXiv:2306.09750 :ramen: tl;dr: Google proposed Federated Learning (FL) as a novel paradigm to train Machine Learning (ML) models across the participants of a federation while preserving data privacy. Since its birth, Centralized FL (CFL) has been the most used approach, where a central entity aggregates participants' models to create a global one. Fedstellar is a novel platform designed to train FL models in a decentralized, semi-decentralized, and centralized fashion across diverse federations of physical or virtualized devices.... :seedling: Read more →
Towards Quantum Federated Learning
:brain: Authors: Chao Ren, Han Yu, Rudai Yan, Minrui Xu, Yuan Shen, Huihui Zhu, Dusit Niyato, Zhao Yang Dong, Leong Chuan Kwek :paw_prints: Details:arXiv:2306.09912 :ramen: tl;dr: Quantum Federated Learning (QFL) is an emerging interdisciplinary field that merges the principles of Quantum Computing (QC) and Federated learning (FL) We aim to provide a comprehensive understanding of the principles, techniques, and emerging applications of QFL. This review serves as a first-of-its-kind comprehensive guide for researchers.... :sunflower: Comments: Survey of quantum federated learning (QFL) :seedling: Read more →
HePCo: Data-Free Heterogeneous Prompt Consolidation for Continual Federated Learning
:brain: Authors: Shaunak Halbe, James Seale Smith, Junjiao Tian, Zsolt Kira :paw_prints: Details:arXiv:2306.09970 :ramen: tl;dr: In this paper, we focus on the important yet understudied problem of Continual Federated Learning. Existing attempts at this problem tend to impose large overheads on clients and communication channels. We propose a novel and lightweight generation and distillation scheme to consolidate client models at the server.... :seedling: Read more →
This is an autogenerated issue. It uses the facebook/bart-large-cnn model to generate summaries for each papers appeared below.
Keyword: IoT (6 papers)
Spatiotemporal-Augmented Graph Neural Networks for Human Mobility Simulation
:brain: Authors: Yu Wang, Tongya Zheng, Shunyu Liu, Kaixuan Chen, Zunlei Feng, Yunzhi Hao, Mingli Song:paw_prints: Details: arXiv:2306.09381
:ramen: tl;dr: Human mobility patterns have shown significant applications in policy-decision scenarios and economic behavior researches. Existing methods mostly rely on the static relationships of locations. We propose a novel framework to model the dynamic spatiotemporal effects of locations, namely SpatioTemporal-Augmented gRaph neural networks (STAR)...
:seedling: Read more →
Adaptive Hierarchical SpatioTemporal Network for Traffic Forecasting
:brain: Authors: Yirong Chen, Ziyue Li, Wanli Ouyang, Michael Lepech:paw_prints: Details: arXiv:2306.09386
:ramen: tl;dr: Accurate traffic forecasting is vital to intelligent transportation systems, which are widely adopted to solve urban traffic issues. Existing traffic forecasting studies focus on modeling spatial-temporal dynamics in traffic data. In this work, we propose an Adaptive Hierarchical SpatioTemporal Network (AHSTN) to promote traffic forecasting....
:sunflower: Comments: Accepted to IEEE CASE 2023
:seedling: Read more →
Multi-Objective and Model-Predictive Tree Search for Spatiotemporal Informative Planning
:brain: Authors: Weizhe Chen, Lantao Liu:paw_prints: Details: arXiv:2306.09608
:ramen: tl;dr: Adaptive sampling and planning in robotic environmental monitoring are challenging when the target environmental process varies over space and time. We propose a Monte Carlo tree search method which not only well balances the environment exploration and exploitation in space, but also catches up to the temporal environmental dynamics. This is achieved by incorporating multi-objective optimization and a look-ahead model-predictive rewarding mechanism....
:sunflower: Comments: Accepted to the 58th IEEE Conference on Decision and Control (CDC 2019)
:seedling: Read more →
EVOPOSE: A Recursive Transformer For 3D Human Pose Estimation With Kinematic Structure Priors
:brain: Authors: Yaqi Zhang, Yan Lu, Bin Liu, Zhiwei Zhao, Qi Chu, Nenghai Yu:paw_prints: Details: arXiv:2306.09615
:ramen: tl;dr: Transformer is popular in recent 3D human pose estimation, which utilizes long-term modeling to lift 2D keypoints into the 3D space. Current transformer-based methods do not fully exploit the prior knowledge of the human skeleton provided by the kinematic structure. In this paper, we propose a novel transformer- based model EvoPose to introduce the human body prior knowledge....
:sunflower: Comments: 5 pages, 2 figures, 4 tables, published in the proceedings of IEEE ICASSP 2023
:seedling: Read more →
Dynamic Decision Tree Ensembles for Energy-Efficient Inference on IoT Edge Nodes
:brain: Authors: Francesco Daghero, Alessio Burrello, Enrico Macii, Paolo Montuschi, Massimo Poncino, Daniele Jahier Pagliari:paw_prints: Details: arXiv:2306.09789
:ramen: tl;dr: With the increasing popularity of Internet of Things (IoT) devices, there is a growing need for energy-efficient Machine Learning (ML) models that can run on constrained edge nodes. Decision tree ensembles, such as Random Forests (RFs) and Gradient Boosting (GBTs), are particularly suited for this task....
:sunflower: Comments: This article has been accepted for publication in IEEE Internet of Things Journal
:seedling: Read more →
PanoOcc: Unified Occupancy Representation for Camera-based 3D Panoptic Segmentation
:brain: Authors: Yuqi Wang, Yuntao Chen, Xingyu Liao, Lue Fan, Zhaoxiang Zhang:paw_prints: Details: arXiv:2306.10013
:ramen: tl;dr: PanoOcc uses voxel queries to aggregate spatiotemporal information from multi-frame and multi-view images in a coarse-to-fine scheme. It integrates feature learning and scene representation into a unified occupancy representation. The code will be released at https //github.com/Robertwyq/Pano Occ....
:sunflower: Comments: technical report
:seedling: Read more →
Keyword: edge computing (3 papers)
A flexible algorithm to offload DAG applications for edge computing
:brain: Authors: Gabriel F. C. de Queiroz, José F. de Rezende, Valmir C. Barbosa:paw_prints: Details: arXiv:2306.09458
:ramen: tl;dr: Multi-access Edge Computing (MEC) is an enabling technology to leverage new network applications, such as virtual/augmented reality. MEC provides faster task processing at the network edge. This is done by deploying servers closer to the end users to run the network applications....
:seedling: Read more →
A Design Approach and Prototype Implementation for Factory Monitoring Based on Virtual and Augmented Reality at the Edge of Industry 4.0
:brain: Authors: Christos Anagnostopoulos, Georgios Mylonas, Apostolos P. Fournaris, Christos Koulamas:paw_prints: Details: arXiv:2306.09692
:ramen: tl;dr: We present our approach for a monitoring intervention inside a factory space utilizing both VR and AR. We discuss its main design directions, as well as a basic ontology to aid in simple description of factory assets. In order to highlight the design aspects of our approach, we present a prototype implementation, based on a use case scenario in a factory site....
:sunflower: Comments: Preprint submitted to the 21st IEEE International Conference on Industrial Informatics, INDIN'23
:seedling: Read more →
Can Orbital Servers Provide Mars-Wide Edge Computing?
:brain: Authors: Tobias Pfandzelter, David Bermbach:paw_prints: Details: arXiv:2306.09756
:ramen: tl;dr: Human landing, exploration and settlement on Mars will require local compute resources at the Mars edge. Landing such resources on Mars is an expensive endeavor. Instead, concepts from low-Earth orbit edge computing may be applied to Mars edge computing. This could lower launching costs of compute resources for Mars while also providing Mars-wide networking....
:seedling: Read more →
Keyword: Blockchain (2 papers)
PIEChain -- A Practical Blockchain Interoperability Framework
:brain: Authors: Daniël Reijsbergen, Aung Maw, Jingchi Zhang, Tien Tuan Anh Dinh, Anwitaman Datta:paw_prints: Details: arXiv:2306.09735
:ramen: tl;dr: PIEChain is a general, Kafka-based cross-chain communication framework. We utilize PIEChain for a practical case study a cross- Chain auction. Users who hold tokens on multiple chains bid for a ticket sold on another chain. PIE chain is the first publicly available, practical implementation of a general framework for cross-Chain communication....
:seedling: Read more →
CroCoDai: A Stablecoin for Cross-Chain Commerce
:brain: Authors: Daniël Reijsbergen, Bretislav Hajek, Tien Tuan Anh Dinh, Jussi Keppo, Hank Korth, Anwitaman Datta:paw_prints: Details: arXiv:2306.09754
:ramen: tl;dr: Decentralized Finance (DeFi) has grown rapidly in value in recent years. The global DeFi ecosystem is fragmented into multiple blockchains, fueling the demand for cross-chain commerce. Stablecoins, which are pegged to a non-volatile asset such as the US dollar, help mitigate the risk associated with price fluctuations....
:seedling: Read more →
Keyword: Federated Learning (7 papers)
A Simple Data Augmentation for Feature Distribution Skewed Federated Learning
:brain: Authors: Yunlu Yan, Lei Zhu:paw_prints: Details: arXiv:2306.09363
:ramen: tl;dr: Federated learning (FL) facilitates collaborative learning among multiple clients. Main challenge lies in the feature shift caused by the different underlying distributions of local datasets. FedRDN is a plug-and-play component, which can be seamlessly integrated into the data augmentation flow with only a few lines of code....
:sunflower: Comments: 11 pages, 3 figures
:seedling: Read more →
Towards Practical Federated Causal Structure Learning
:brain: Authors: Zhaoyu Wang, Pingchuan Ma, Shuai Wang:paw_prints: Details: arXiv:2306.09433
:ramen: tl;dr: The process of causal structure learning involves identifying causal graphs from observational data. FedC2SL is a federated constraint-based causal structurelearning scheme that learns causal graphs. It examines conditional independence between two variables under a condition set without collecting raw data from clients. The study finds it demonstrates encouraging performance and strong resilience to data heterogeneity among clients....
:seedling: Read more →
Opportunistic Transmission of Distributed Learning Models in Mobile UAVs
:brain: Authors: Jingxin Li, Xiaolan Liu, Toktam Mahmoodi:paw_prints: Details: arXiv:2306.09484
:ramen: tl;dr: Proposal aims to take advantage of the proximity of users to the base station or the general condition of the wireless transmission channel. During the training, intermediate model parameters are uploaded to the server, opportunistically and based on the wireless channel condition. Test accuracy increases by up to 13.47% with just one round of extra transmission....
:seedling: Read more →
FedMultimodal: A Benchmark For Multimodal Federated Learning
:brain: Authors: Tiantian Feng, Digbalay Bose, Tuo Zhang, Rajat Hebbar, Anil Ramakrishna, Rahul Gupta, Mi Zhang, Salman Avestimehr, Shrikanth Narayanan:paw_prints: Details: arXiv:2306.09486
:ramen: tl;dr: Federated Learning (FL) has become an emerging machine learning technique to tackle data privacy challenges through collaborative training. Despite significant efforts that have been made to FL in fields like computer vision, audio, and natural language processing, the FL applications utilizing multimodal data streams remain largely unexplored....
:sunflower: Comments: This paper was accepted to KDD 2023 Applied Data Science (ADS) track
:seedling: Read more →
Fedstellar: A Platform for Decentralized Federated Learning
:brain: Authors: Enrique Tomás Martínez Beltrán, Ángel Luis Perales Gómez, Chao Feng, Pedro Miguel Sánchez Sánchez, Sergio López Bernal, Gérôme Bovet, Manuel Gil Pérez, Gregorio Martínez Pérez, Alberto Huertas Celdrán:paw_prints: Details: arXiv:2306.09750
:ramen: tl;dr: Google proposed Federated Learning (FL) as a novel paradigm to train Machine Learning (ML) models across the participants of a federation while preserving data privacy. Since its birth, Centralized FL (CFL) has been the most used approach, where a central entity aggregates participants' models to create a global one. Fedstellar is a novel platform designed to train FL models in a decentralized, semi-decentralized, and centralized fashion across diverse federations of physical or virtualized devices....
:seedling: Read more →
Towards Quantum Federated Learning
:brain: Authors: Chao Ren, Han Yu, Rudai Yan, Minrui Xu, Yuan Shen, Huihui Zhu, Dusit Niyato, Zhao Yang Dong, Leong Chuan Kwek:paw_prints: Details: arXiv:2306.09912
:ramen: tl;dr: Quantum Federated Learning (QFL) is an emerging interdisciplinary field that merges the principles of Quantum Computing (QC) and Federated learning (FL) We aim to provide a comprehensive understanding of the principles, techniques, and emerging applications of QFL. This review serves as a first-of-its-kind comprehensive guide for researchers....
:sunflower: Comments: Survey of quantum federated learning (QFL)
:seedling: Read more →
HePCo: Data-Free Heterogeneous Prompt Consolidation for Continual Federated Learning
:brain: Authors: Shaunak Halbe, James Seale Smith, Junjiao Tian, Zsolt Kira:paw_prints: Details: arXiv:2306.09970
:ramen: tl;dr: In this paper, we focus on the important yet understudied problem of Continual Federated Learning. Existing attempts at this problem tend to impose large overheads on clients and communication channels. We propose a novel and lightweight generation and distillation scheme to consolidate client models at the server....
:seedling: Read more →