This is an autogenerated issue. It uses the facebook/bart-large-cnn model to generate summaries for each papers appeared below.
Keyword: IoT (8 papers)
Spatiotemporal Transformer for Stock Movement Prediction
:brain: Authors: Daniel Boyle, Jugal Kalita :paw_prints: Details:arXiv:2305.03835 :ramen: tl;dr: The dynamic, non-linear nature of financial markets makes it extremely hard to predict future price movements. Correlating these factors is virtually impossible for a human. We propose a novel approach using a Spatiotemporal Transformer-LSTM model for stock movement prediction. Our model obtains accuracies of 63.707 and 56.879 percent against the ACL18 and KDD17 datasets.... :seedling: Read more →
Rateless Coded Blockchain for Dynamic IoT Networks
:brain: Authors: Changlin Yang, Alexei Ashikhmin, Xiaodong Wang, Zibin Zheng :paw_prints: Details:arXiv:2305.03895 :ramen: tl;dr: A key constraint that limits the implementation of blockchain in Internet of Things (IoT) is its large storage requirement. This increases the burden on blockchain nodes, and increases the communication overhead for new nodes joining the network. In order to reduce storage requirements without compromising on system security and integrity, coded blockchains have been recently proposed.... :seedling: Read more →
Memory Disaggregation: Advances and Open Challenges
:brain: Authors: Hasan Al Maruf, Mosharaf Chowdhury :paw_prints: Details:arXiv:2305.03943 :ramen: tl;dr: Compute and memory are tightly coupled within each server in traditional datacenters. Large-scale datacenter operators have identified this coupling as a root cause behind fleet-wide resource underutilization. With the advent of ultra-fast networks and cache-coherent interfaces, memory disaggregation has emerged as a potential solution.... :seedling: Read more →
LibAM: An Area Matching Framework for Detecting Third-party Libraries in Binaries
:brain: Authors: Siyuan Li, Yongpan Wang, Chaopeng Dong, Shouguo Yang, Hong Li, Hao Sun, Zhe Lang, Zuxin Chen, Weijie Wang, Hongsong Zhu, Limin Sun :paw_prints: Details:arXiv:2305.04026 :ramen: tl;dr: Third-party libraries (TPLs) are extensively utilized by developers to expedite the software development process and incorporate external functionalities. insecure TPL reuse can lead to significant security risks. Existing methods are employed to determine the presence of TPL code in the target binary. We introduce LibAM, a novel Area Matching framework that connects isolated functions into function areas on FCG.... :sunflower: Comments: 33 pages, 12 figures :seedling: Read more →
Blockchained Federated Learning for Internet of Things: A Comprehensive Survey
:brain: Authors: Yanna Jiang, Baihe Ma, Xu Wang, Ping Yu, Guangsheng Yu, Zhe Wang, Wei Ni, Ren Ping Liu :paw_prints: Details:arXiv:2305.04513 :ramen: tl;dr: The demand for intelligent industries and smart services based on big data is rising rapidly with the increasing digitization and intelligence of the modern world. This survey comprehensively reviews Blockchained Federated Learning (BlockFL) The features of decentralization and transparency make BlockFL a secure and effective solution for distributed model training.... :seedling: Read more →
Physical Layer Secret Key Generation with Kalman Filter Detrending
:brain: Authors: Miroslav Mitev, Arsenia Chorti, Gerhard Fettweis :paw_prints: Details:arXiv:2305.04540 :ramen: tl;dr: Secret key generation (SKG) from channel state information (CSI) during the channel's coherence time. Key must be generated using only the random and unpredictable part of CSI. To detrend CSI measurements from deterministic components, a simple and lightweight approach based on Kalman filters is proposed.... :seedling: Read more →
Joint Task Offloading and Resource Allocation for Streaming Application in Cooperative Mobile Edge Computing
:brain: Authors: Xiang Li, Rongfei Fan, Han Hu, Xiangming Li :paw_prints: Details:arXiv:2305.04747 :ramen: tl;dr: Mobile edge computing (MEC) enables resource-limited IoT devices to complete computation-intensive or delay-sensitive task. Due to channel occlusion, some users may not be able to access the computation capability directly from the base station. Many other devices in the MEC system can serve as cooperative nodes to collect the tasks of these users.... :sunflower: Comments: 29 pages,9 figures,1 table :seedling: Read more →
Cheshire: A Lightweight, Linux-Capable RISC-V Host Platform for Domain-Specific Accelerator Plug-In
:brain: Authors: Alessandro Ottaviano, Thomas Benz, Paul Scheffler, Luca Benini :paw_prints: Details:arXiv:2305.04760 :ramen: tl;dr: Cheshire is a lightweight and modular 64-bit Linux-capable host platform designed for the seamless plug-in of domain-specific accelerators. It features a unique low-pin-count DRAM interface, a last-level cache configurable as scratchpad memory, and a DMA engine.... :sunflower: Comments: 5 pages, 11 figures, submitted to an IEEE journal for possible publication :seedling: Read more →
Keyword: Blockchain (7 papers)
DORA: Distributed Oracle Agreement with Simple Majority
:brain: Authors: Prasanth Chakka, Saurabh Joshi, Aniket Kate, Joshua Tobkin, David Yang :paw_prints: Details:arXiv:2305.03903 :ramen: tl;dr: Oracle networks feeding off-chain information to a blockchain are required to solve a distributed agreement problem. These networks receive information from multiple sources and at different times. We define a notion of agreement distance and leverage the availability of a state machine replication (SMR) service. This allows the oracle network to operate much more efficiently and horizontally scale.... :sunflower: Comments: 13 pages, 4 figures, accepted at ICDCS 2023 :seedling: Read more →
An Overview of AI and Blockchain Integration for Privacy-Preserving
:brain: Authors: Zongwei Li, Dechao Kong, Yuanzheng Niu, Hongli Peng, Xiaoqi Li, Wenkai Li :paw_prints: Details:arXiv:2305.03928 :ramen: tl;dr: Privacy protection techniques arising from AI and blockchain integration are of notable significance. These techniques also guarantee security and dependability of data. This research classifies and summarizes privacy protection techniques based on AI-blockchain application scenarios and technical schemes. In conclusion, this paper outlines the future directions of privacy protection technologies emerging from AI.... :sunflower: Comments: 23 pages, 4 figures :seedling: Read more →
A Blockchain-based Platform for Reliable Inference and Training of Large-Scale Models
:brain: Authors: Sanghyeon Park, Junmo Lee, Soo-Mook Moon :paw_prints: Details:arXiv:2305.04062 :ramen: tl;dr: BRAIN harnesses a unique two-phase transaction mechanism. It allows real-time processing via pipelining by separating request and response transactions. BRAIN's tasks-per-second performance is 454.4293 times greater than that of a naive single-phase implementation.... :sunflower: Comments: 12 pages, 2 figures :seedling: Read more →
An Empirical Study on Governance in Bitcoin's Consensus Evolution
:brain: Authors: Jakob Svennevik Notland, Mariusz Nowostawski, Jingyue Li :paw_prints: Details:arXiv:2305.04079 :ramen: tl;dr: Blockchain systems run consensus rules as code to agree on the state of the distributed ledger and secure the network. Changing these rules can be risky and challenging. Arguably, Bitcoin has seen centralisation tendencies in pools and in development. However, how these tendencies influence blockchain governance has received minimal community and academic attention.... :seedling: Read more →
Stackelberg Attacks or: How I Learned to Stop Worrying and Trust the Blockchain
:brain: Authors: Daji Landis, Nikolaj I. Schwartzbach :paw_prints: Details:arXiv:2305.04373 :ramen: tl;dr: A game that remains secure when the agents can deploy contracts is said to be Stackelberg resilient. We demonstrate the non-triviality of side contract resilience by analyzing two smart contracts for decentralized commerce. Our work highlights an issue that is necessary to address to ensure the secure deployment of smart contracts.... :sunflower: Comments: Updated version of 2301.08523. arXiv admin note: substantial text overlap with arXiv:2301.08523 :seedling: Read more →
Capturing Smart Contract Design with DCR Graphs
:brain: Authors: Mojtaba Eshghie, Wolfgang Ahrendt, Cyrille Artho, Thomas Troels Hildebrandt, Gerardo Schneider :paw_prints: Details:arXiv:2305.04581 :ramen: tl;dr: Mainstream smart contract programming languages such as Solidity do not have explicit notions of roles, action dependencies, and time. This makes it very hard to design and analyze smart contracts. DCR graphs are a suitable formalization tool for smart contracts because they explicitly and visually capture these features.... :seedling: Read more →
PBL: System for Creating and Maintaining Personal Blockchain Ledgers
:brain: Authors: Collin Connors, Dilip Sarkar :paw_prints: Details:arXiv:2305.04723 :ramen: tl;dr: Blockchain technology has experienced substantial growth in recent years, yet the diversity of blockchain applications has been limited. This work presents a novel architecture for creating and maintaining personal blockchain ledgers that address these concerns. Unlike traditional blockchain, which stores all transactions of multiple users, our novel personal blockchains are designed to allow individuals to maintain their privacy without extensive technical expertise.... :seedling: Read more →
Keyword: Federated Learning (9 papers)
Data Station: Delegated, Trustworthy, and Auditable Computation to Enable Data-Sharing Consortia with a Data Escrow
:brain: Authors: Siyuan Xia, Zhiru Zhu, Chris Zhu, Jinjin Zhao, Kyle Chard, Aaron J. Elmore, Ian Foster, Michael Franklin, Sanjay Krishnan, Raul Castro Fernandez :paw_prints: Details:arXiv:2305.03842 :ramen: tl;dr: Data Station is a data escrow designed to enable the formation of data-sharing consortia. Data owners share data with the escrow knowing it will not be released without their consent. Data Station leverages hardware enclaves to generate trust among participants, and exploits the centralization of data and computation to generate an audit log.... :seedling: Read more →
Decentralised Semi-supervised Onboard Learning for Scene Classification in Low-Earth Orbit
:brain: Authors: Johan Östman, Pablo Gomez, Vinutha Magal Shreenath, Gabriele Meoni :paw_prints: Details:arXiv:2305.04059 :ramen: tl;dr: Onboard machine learning on the latest satellite hardware offers the potential for significant savings in communication and operational costs. We evaluate mission scenarios employing both decentralised and federated learning approaches. All scenarios achieve convergence to high accuracy (around 91% on EuroSAT RGB dataset) within a one-day mission timeframe.... :sunflower: Comments: Accepted at IAA SSEO 2023 :seedling: Read more →
Exploring One-shot Semi-supervised Federated Learning with A Pre-trained Diffusion Model
:brain: Authors: Mingzhao Yang, Shangchao Su, Bin Li, Xiangyang Xue :paw_prints: Details:arXiv:2305.04063 :ramen: tl;dr: Federated learning is a privacy-preserving collaborative learning approach. FedDISC is a Federated Diffusion Inspired Semi-supervised Co-training method. It uses a pre-trained diffusion model to conditionally generate pseudo-samples complying with the client distributions. Our method does not require local training and only involves forward inference on the clients.... :seedling: Read more →
Portfolio-Based Incentive Mechanism Design for Cross-Device Federated Learning
:brain: Authors: Jiaxi Yang, Sheng Cao, Cuifang Zhao, Weina Niu, Li-Chuan Tsai :paw_prints: Details:arXiv:2305.04081 :ramen: tl;dr: There has been a significant increase in attention towards designing incentive mechanisms for federated learning (FL) In this paper, we focus on cross-device FL and propose a multi-level FL architecture under the real scenarios. We propose FL Incentive Mechanism based on Portfolio theory (FL-IMP)... :seedling: Read more →
Gradient Leakage Defense with Key-Lock Module for Federated Learning
:brain: Authors: Hanchi Ren, Jingjing Deng, Xianghua Xie, Xiaoke Ma, Jianfeng Ma :paw_prints: Details:arXiv:2305.04095 :ramen: tl;dr: Federated Learning (FL) is a widely adopted privacy-preserving machine learning approach where private data remains local. However, recent findings reveal that privacy may be compromised and sensitive information potentially recovered from shared gradients. In this study, we offer detailed analysis and a novel perspective on understanding the gradient leakage problem.... :sunflower: Comments: The source code can be found at this https URL :seedling: Read more →
MrTF: Model Refinery for Transductive Federated Learning
:brain: Authors: Xin-Chun Li, Yang Yang, De-Chuan Zhan :paw_prints: Details:arXiv:2305.04201 :ramen: tl;dr: Current learning paradigms are devoted to solving the data heterogeneity problem without considering the to-be-inferred data. We propose a novel learning paradigm named transductive federated learning (TFL) We propose several techniques including stabilized teachers, rectified distillation, and clustered label refinery to facilitate the model refinery process.... :sunflower: Comments: Minor Revision to DMKD Journal :seedling: Read more →
Towards Achieving Near-optimal Utility for Privacy-Preserving Federated Learning via Data Generation and Parameter Distortion
:brain: Authors: Xiaojin Zhang, Kai Chen, Qiang Yang :paw_prints: Details:arXiv:2305.04288 :ramen: tl;dr: Federated learning (FL) enables participating parties to collaboratively build a global model with boosted utility without disclosing private data information. We want to identify under what general conditions privacy-preserving federated learning can achieve near-optimal utility via data generation and parameter distortion. We provide an upper bound for the trade-off between privacy and utility.... :seedling: Read more →
FedZKP: Federated Model Ownership Verification with Zero-knowledge Proof
:brain: Authors: Wenyuan Yang, Yuguo Yin, Gongxi Zhu, Hanlin Gu, Lixin Fan, Xiaochun Cao, Qiang Yang :paw_prints: Details:arXiv:2305.04507 :ramen: tl;dr: Federated learning (FL) allows multiple parties to cooperatively learn a federated model without sharing private data with each other. The need of protecting such federated models from being plagiarized or misused, therefore, motivates us to propose a provable secure model ownership verification scheme using zero-knowledge proof.... :seedling: Read more →
Federated Learning in Wireless Networks via Over-the-Air Computations
:brain: Authors: Halil Yigit Oksuz, Fabio Molinari, Henning Sprekeler, Jörg Raisch :paw_prints: Details:arXiv:2305.04630 :ramen: tl;dr: In a multi-agent system, agents can cooperatively learn a model from data by exchanging their estimated model parameters. Over-the-Air Computation exploits the interference property of the wireless channel. Standard communication schemes prevent interference by enabling transmissions of signals from different agents at distinct time or frequency slots.... :sunflower: Comments: 8 pages, 2 figures, submitted to 62nd IEEE Conference on Decision and Control :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 (8 papers)
Spatiotemporal Transformer for Stock Movement Prediction
:brain: Authors: Daniel Boyle, Jugal Kalita:paw_prints: Details: arXiv:2305.03835
:ramen: tl;dr: The dynamic, non-linear nature of financial markets makes it extremely hard to predict future price movements. Correlating these factors is virtually impossible for a human. We propose a novel approach using a Spatiotemporal Transformer-LSTM model for stock movement prediction. Our model obtains accuracies of 63.707 and 56.879 percent against the ACL18 and KDD17 datasets....
:seedling: Read more →
Rateless Coded Blockchain for Dynamic IoT Networks
:brain: Authors: Changlin Yang, Alexei Ashikhmin, Xiaodong Wang, Zibin Zheng:paw_prints: Details: arXiv:2305.03895
:ramen: tl;dr: A key constraint that limits the implementation of blockchain in Internet of Things (IoT) is its large storage requirement. This increases the burden on blockchain nodes, and increases the communication overhead for new nodes joining the network. In order to reduce storage requirements without compromising on system security and integrity, coded blockchains have been recently proposed....
:seedling: Read more →
Memory Disaggregation: Advances and Open Challenges
:brain: Authors: Hasan Al Maruf, Mosharaf Chowdhury:paw_prints: Details: arXiv:2305.03943
:ramen: tl;dr: Compute and memory are tightly coupled within each server in traditional datacenters. Large-scale datacenter operators have identified this coupling as a root cause behind fleet-wide resource underutilization. With the advent of ultra-fast networks and cache-coherent interfaces, memory disaggregation has emerged as a potential solution....
:seedling: Read more →
LibAM: An Area Matching Framework for Detecting Third-party Libraries in Binaries
:brain: Authors: Siyuan Li, Yongpan Wang, Chaopeng Dong, Shouguo Yang, Hong Li, Hao Sun, Zhe Lang, Zuxin Chen, Weijie Wang, Hongsong Zhu, Limin Sun:paw_prints: Details: arXiv:2305.04026
:ramen: tl;dr: Third-party libraries (TPLs) are extensively utilized by developers to expedite the software development process and incorporate external functionalities. insecure TPL reuse can lead to significant security risks. Existing methods are employed to determine the presence of TPL code in the target binary. We introduce LibAM, a novel Area Matching framework that connects isolated functions into function areas on FCG....
:sunflower: Comments: 33 pages, 12 figures
:seedling: Read more →
Blockchained Federated Learning for Internet of Things: A Comprehensive Survey
:brain: Authors: Yanna Jiang, Baihe Ma, Xu Wang, Ping Yu, Guangsheng Yu, Zhe Wang, Wei Ni, Ren Ping Liu:paw_prints: Details: arXiv:2305.04513
:ramen: tl;dr: The demand for intelligent industries and smart services based on big data is rising rapidly with the increasing digitization and intelligence of the modern world. This survey comprehensively reviews Blockchained Federated Learning (BlockFL) The features of decentralization and transparency make BlockFL a secure and effective solution for distributed model training....
:seedling: Read more →
Physical Layer Secret Key Generation with Kalman Filter Detrending
:brain: Authors: Miroslav Mitev, Arsenia Chorti, Gerhard Fettweis:paw_prints: Details: arXiv:2305.04540
:ramen: tl;dr: Secret key generation (SKG) from channel state information (CSI) during the channel's coherence time. Key must be generated using only the random and unpredictable part of CSI. To detrend CSI measurements from deterministic components, a simple and lightweight approach based on Kalman filters is proposed....
:seedling: Read more →
Joint Task Offloading and Resource Allocation for Streaming Application in Cooperative Mobile Edge Computing
:brain: Authors: Xiang Li, Rongfei Fan, Han Hu, Xiangming Li:paw_prints: Details: arXiv:2305.04747
:ramen: tl;dr: Mobile edge computing (MEC) enables resource-limited IoT devices to complete computation-intensive or delay-sensitive task. Due to channel occlusion, some users may not be able to access the computation capability directly from the base station. Many other devices in the MEC system can serve as cooperative nodes to collect the tasks of these users....
:sunflower: Comments: 29 pages,9 figures,1 table
:seedling: Read more →
Cheshire: A Lightweight, Linux-Capable RISC-V Host Platform for Domain-Specific Accelerator Plug-In
:brain: Authors: Alessandro Ottaviano, Thomas Benz, Paul Scheffler, Luca Benini:paw_prints: Details: arXiv:2305.04760
:ramen: tl;dr: Cheshire is a lightweight and modular 64-bit Linux-capable host platform designed for the seamless plug-in of domain-specific accelerators. It features a unique low-pin-count DRAM interface, a last-level cache configurable as scratchpad memory, and a DMA engine....
:sunflower: Comments: 5 pages, 11 figures, submitted to an IEEE journal for possible publication
:seedling: Read more →
Keyword: Blockchain (7 papers)
DORA: Distributed Oracle Agreement with Simple Majority
:brain: Authors: Prasanth Chakka, Saurabh Joshi, Aniket Kate, Joshua Tobkin, David Yang:paw_prints: Details: arXiv:2305.03903
:ramen: tl;dr: Oracle networks feeding off-chain information to a blockchain are required to solve a distributed agreement problem. These networks receive information from multiple sources and at different times. We define a notion of agreement distance and leverage the availability of a state machine replication (SMR) service. This allows the oracle network to operate much more efficiently and horizontally scale....
:sunflower: Comments: 13 pages, 4 figures, accepted at ICDCS 2023
:seedling: Read more →
An Overview of AI and Blockchain Integration for Privacy-Preserving
:brain: Authors: Zongwei Li, Dechao Kong, Yuanzheng Niu, Hongli Peng, Xiaoqi Li, Wenkai Li:paw_prints: Details: arXiv:2305.03928
:ramen: tl;dr: Privacy protection techniques arising from AI and blockchain integration are of notable significance. These techniques also guarantee security and dependability of data. This research classifies and summarizes privacy protection techniques based on AI-blockchain application scenarios and technical schemes. In conclusion, this paper outlines the future directions of privacy protection technologies emerging from AI....
:sunflower: Comments: 23 pages, 4 figures
:seedling: Read more →
A Blockchain-based Platform for Reliable Inference and Training of Large-Scale Models
:brain: Authors: Sanghyeon Park, Junmo Lee, Soo-Mook Moon:paw_prints: Details: arXiv:2305.04062
:ramen: tl;dr: BRAIN harnesses a unique two-phase transaction mechanism. It allows real-time processing via pipelining by separating request and response transactions. BRAIN's tasks-per-second performance is 454.4293 times greater than that of a naive single-phase implementation....
:sunflower: Comments: 12 pages, 2 figures
:seedling: Read more →
An Empirical Study on Governance in Bitcoin's Consensus Evolution
:brain: Authors: Jakob Svennevik Notland, Mariusz Nowostawski, Jingyue Li:paw_prints: Details: arXiv:2305.04079
:ramen: tl;dr: Blockchain systems run consensus rules as code to agree on the state of the distributed ledger and secure the network. Changing these rules can be risky and challenging. Arguably, Bitcoin has seen centralisation tendencies in pools and in development. However, how these tendencies influence blockchain governance has received minimal community and academic attention....
:seedling: Read more →
Stackelberg Attacks or: How I Learned to Stop Worrying and Trust the Blockchain
:brain: Authors: Daji Landis, Nikolaj I. Schwartzbach:paw_prints: Details: arXiv:2305.04373
:ramen: tl;dr: A game that remains secure when the agents can deploy contracts is said to be Stackelberg resilient. We demonstrate the non-triviality of side contract resilience by analyzing two smart contracts for decentralized commerce. Our work highlights an issue that is necessary to address to ensure the secure deployment of smart contracts....
:sunflower: Comments: Updated version of 2301.08523. arXiv admin note: substantial text overlap with arXiv:2301.08523
:seedling: Read more →
Capturing Smart Contract Design with DCR Graphs
:brain: Authors: Mojtaba Eshghie, Wolfgang Ahrendt, Cyrille Artho, Thomas Troels Hildebrandt, Gerardo Schneider:paw_prints: Details: arXiv:2305.04581
:ramen: tl;dr: Mainstream smart contract programming languages such as Solidity do not have explicit notions of roles, action dependencies, and time. This makes it very hard to design and analyze smart contracts. DCR graphs are a suitable formalization tool for smart contracts because they explicitly and visually capture these features....
:seedling: Read more →
PBL: System for Creating and Maintaining Personal Blockchain Ledgers
:brain: Authors: Collin Connors, Dilip Sarkar:paw_prints: Details: arXiv:2305.04723
:ramen: tl;dr: Blockchain technology has experienced substantial growth in recent years, yet the diversity of blockchain applications has been limited. This work presents a novel architecture for creating and maintaining personal blockchain ledgers that address these concerns. Unlike traditional blockchain, which stores all transactions of multiple users, our novel personal blockchains are designed to allow individuals to maintain their privacy without extensive technical expertise....
:seedling: Read more →
Keyword: Federated Learning (9 papers)
Data Station: Delegated, Trustworthy, and Auditable Computation to Enable Data-Sharing Consortia with a Data Escrow
:brain: Authors: Siyuan Xia, Zhiru Zhu, Chris Zhu, Jinjin Zhao, Kyle Chard, Aaron J. Elmore, Ian Foster, Michael Franklin, Sanjay Krishnan, Raul Castro Fernandez:paw_prints: Details: arXiv:2305.03842
:ramen: tl;dr: Data Station is a data escrow designed to enable the formation of data-sharing consortia. Data owners share data with the escrow knowing it will not be released without their consent. Data Station leverages hardware enclaves to generate trust among participants, and exploits the centralization of data and computation to generate an audit log....
:seedling: Read more →
Decentralised Semi-supervised Onboard Learning for Scene Classification in Low-Earth Orbit
:brain: Authors: Johan Östman, Pablo Gomez, Vinutha Magal Shreenath, Gabriele Meoni:paw_prints: Details: arXiv:2305.04059
:ramen: tl;dr: Onboard machine learning on the latest satellite hardware offers the potential for significant savings in communication and operational costs. We evaluate mission scenarios employing both decentralised and federated learning approaches. All scenarios achieve convergence to high accuracy (around 91% on EuroSAT RGB dataset) within a one-day mission timeframe....
:sunflower: Comments: Accepted at IAA SSEO 2023
:seedling: Read more →
Exploring One-shot Semi-supervised Federated Learning with A Pre-trained Diffusion Model
:brain: Authors: Mingzhao Yang, Shangchao Su, Bin Li, Xiangyang Xue:paw_prints: Details: arXiv:2305.04063
:ramen: tl;dr: Federated learning is a privacy-preserving collaborative learning approach. FedDISC is a Federated Diffusion Inspired Semi-supervised Co-training method. It uses a pre-trained diffusion model to conditionally generate pseudo-samples complying with the client distributions. Our method does not require local training and only involves forward inference on the clients....
:seedling: Read more →
Portfolio-Based Incentive Mechanism Design for Cross-Device Federated Learning
:brain: Authors: Jiaxi Yang, Sheng Cao, Cuifang Zhao, Weina Niu, Li-Chuan Tsai:paw_prints: Details: arXiv:2305.04081
:ramen: tl;dr: There has been a significant increase in attention towards designing incentive mechanisms for federated learning (FL) In this paper, we focus on cross-device FL and propose a multi-level FL architecture under the real scenarios. We propose FL Incentive Mechanism based on Portfolio theory (FL-IMP)...
:seedling: Read more →
Gradient Leakage Defense with Key-Lock Module for Federated Learning
:brain: Authors: Hanchi Ren, Jingjing Deng, Xianghua Xie, Xiaoke Ma, Jianfeng Ma:paw_prints: Details: arXiv:2305.04095
:ramen: tl;dr: Federated Learning (FL) is a widely adopted privacy-preserving machine learning approach where private data remains local. However, recent findings reveal that privacy may be compromised and sensitive information potentially recovered from shared gradients. In this study, we offer detailed analysis and a novel perspective on understanding the gradient leakage problem....
:sunflower: Comments: The source code can be found at this https URL
:seedling: Read more →
MrTF: Model Refinery for Transductive Federated Learning
:brain: Authors: Xin-Chun Li, Yang Yang, De-Chuan Zhan:paw_prints: Details: arXiv:2305.04201
:ramen: tl;dr: Current learning paradigms are devoted to solving the data heterogeneity problem without considering the to-be-inferred data. We propose a novel learning paradigm named transductive federated learning (TFL) We propose several techniques including stabilized teachers, rectified distillation, and clustered label refinery to facilitate the model refinery process....
:sunflower: Comments: Minor Revision to DMKD Journal
:seedling: Read more →
Towards Achieving Near-optimal Utility for Privacy-Preserving Federated Learning via Data Generation and Parameter Distortion
:brain: Authors: Xiaojin Zhang, Kai Chen, Qiang Yang:paw_prints: Details: arXiv:2305.04288
:ramen: tl;dr: Federated learning (FL) enables participating parties to collaboratively build a global model with boosted utility without disclosing private data information. We want to identify under what general conditions privacy-preserving federated learning can achieve near-optimal utility via data generation and parameter distortion. We provide an upper bound for the trade-off between privacy and utility....
:seedling: Read more →
FedZKP: Federated Model Ownership Verification with Zero-knowledge Proof
:brain: Authors: Wenyuan Yang, Yuguo Yin, Gongxi Zhu, Hanlin Gu, Lixin Fan, Xiaochun Cao, Qiang Yang:paw_prints: Details: arXiv:2305.04507
:ramen: tl;dr: Federated learning (FL) allows multiple parties to cooperatively learn a federated model without sharing private data with each other. The need of protecting such federated models from being plagiarized or misused, therefore, motivates us to propose a provable secure model ownership verification scheme using zero-knowledge proof....
:seedling: Read more →
Federated Learning in Wireless Networks via Over-the-Air Computations
:brain: Authors: Halil Yigit Oksuz, Fabio Molinari, Henning Sprekeler, Jörg Raisch:paw_prints: Details: arXiv:2305.04630
:ramen: tl;dr: In a multi-agent system, agents can cooperatively learn a model from data by exchanging their estimated model parameters. Over-the-Air Computation exploits the interference property of the wireless channel. Standard communication schemes prevent interference by enabling transmissions of signals from different agents at distinct time or frequency slots....
:sunflower: Comments: 8 pages, 2 figures, submitted to 62nd IEEE Conference on Decision and Control
:seedling: Read more →