Filters








1,697 Hits in 7.7 sec

One for One, or All for All: Equilibria and Optimality of Collaboration in Federated Learning [article]

Avrim Blum, Nika Haghtalab, Richard Lanas Phillips, Han Shao
2021 arXiv   pre-print
In recent years, federated learning has been embraced as an approach for bringing about collaboration across large populations of learning agents.  ...  Furthermore, we compare the sample complexity of incentive-aware collaboration with that of optimal collaboration when one ignores agents' incentives.  ...  In particular, we show that there is a factor Ω( √ k) gap between the socially optimal sample complexity and that of optimal stable or envy-free equilibria for k agents.  ... 
arXiv:2103.03228v1 fatcat:g5q2ionfqndw7ctmivwitgewie

Incentivizing Data Contribution in Cross-Silo Federated Learning [article]

Chao Huang, Huanle Zhang, Xin Liu
2022 arXiv   pre-print
In cross-silo federated learning, clients (e.g., organizations) collectively train a global model using their local data.  ...  We consider two types of equilibrium structures: symmetric and asymmetric equilibria. We show that the three mechanisms admit an identical symmetric equilibrium structure.  ...  INTRODUCTION Federated learning (FL) is a decentralized machine learning paradigm where multiple clients collaboratively train a global model under the orchestration of a central server [1] .  ... 
arXiv:2203.03885v1 fatcat:4pwskfk6drespbgplymfz2rzoq

Modern Applied Science, Vol. 3, No. 1, January 2009, all in one file

Editor MAS
2008 Modern Applied Science  
Yamin, Universiti Kebangsaan Malaysia for helping in doing microanalysis; as well as Desy Anggraini, Chandra Rini P. and Arif Dwi R. who help us in the preparation of the samples.  ...  Acknowledgments The authors would like to thank to The Directorate of Research and Services, Directorate General of Higher Education, The Ministry of Cultural and Education of Republic of Indonesia that  ...  Land use is one of the essential tools for nearly all land development.  ... 
doi:10.5539/mas.v3n1p0 fatcat:jovqztw2qre4ra46ggzarwfatu

Stackelberg Security Game For Optimizing Security Of Federated Internet Of Things Platform Instances

Violeta Damjanovic-Behrendt
2017 Zenodo  
This paper presents an approach for optimal cyber security decisions to protect instances of a federated Internet of Things (IoT) platform in the cloud.  ...  Naïve Q-Learning belongs to the category of active and model-free Machine Learning (ML) techniques in which the agent (either the defender or the attacker) attempts to find an optimal security solution  ...  Fig. 1 1 NIMBLE federated collaboration network with sector-centric platform instances Stackelberg Security Game for Optimizing Security of Federated Internet of Things Platform Instances International  ... 
doi:10.5281/zenodo.1130143 fatcat:ioupac6cbvaz3lftjcptnz5aru

A Flat World, a Level Playing Field, a Small World After All, or None of the Above? A Review of Thomas L. Friedman'sThe World is Flat

Edward E Leamer
2007 Journal of Economic Literature  
Of course, standardization, mechanization, and computerization all work to increase the number of footloose tasks, but innovation and education work in the opposite direction, creating relationship-based  ...  In fact, most of the footloose relationship-free jobs in apparel and footwear and consumer electronics departed the United States several decades ago, and few U.S. workers today feel the force of Chinese  ...  trade, 6 percent in the federal government, 6 percent in hospitals, 5 percent in ambulatory care and so on.  ... 
doi:10.1257/jel.45.1.83 fatcat:ckjoalvntrajfchsfuybzmfdw4

More Than Privacy: Applying Differential Privacy in Key Areas of Artificial Intelligence [article]

Tianqing Zhu and Dayong Ye and Wei Wang and Wanlei Zhou and Philip S. Yu
2020 arXiv   pre-print
With a focus on regular machine learning, distributed machine learning, deep learning, and multi-agent systems, the purpose of this article is to deliver a new view on many possibilities for improving  ...  It can also be used to improve security, stabilize learning, build fair models, and impose composition in selected areas of AI.  ...  Differential privacy in federated learning 1) Overview of federated learning: Federated learning enables individual users to collaboratively learn a shared prediction model while keeping all the training  ... 
arXiv:2008.01916v1 fatcat:ujmxv7eq6jcppndfu5shbzkdom

More Than Privacy: Applying Differential Privacy in Key Areas of Artificial Intelligence

Tianqing Zhu, Dayong Ye, Wei Wang, Wanlei Zhou, Philip Yu
2020 IEEE Transactions on Knowledge and Data Engineering  
With a focus on regular machine learning, distributed machine learning, deep learning, and multi-agent systems, the purpose of this article is to deliver a new view on many possibilities for improving  ...  It can also be used to improve security, stabilize learning, build fair models, and impose composition in selected areas of AI.  ...  Fig. 5 . 5 Differential privacy in deep learning Federated learning enables individual users to collaboratively learn a shared prediction model while keeping all the TABLE 1 1 Properties of differential  ... 
doi:10.1109/tkde.2020.3014246 fatcat:33rl6jxy5rgexpnuel5rvlkg5a

Collaboration in Participant-Centric Federated Learning: A Game-Theoretical Perspective [article]

Guangjing Huang and Xu Chen and Tao Ouyang and Qian Ma and Lin Chen and Junshan Zhang
2022 arXiv   pre-print
Federated learning (FL) is a promising distributed framework for collaborative artificial intelligence model training while protecting user privacy.  ...  A bootstrapping component that has attracted significant research attention is the design of incentive mechanism to stimulate user collaboration in FL.  ...  Summarizing concept and applications in federated learning, Yang et al. classify federated learning into three types: horizontal federated learning, vertical federated learning and federated transfer learning  ... 
arXiv:2207.12030v1 fatcat:6r4task245drzbj5ssknmrfna4

California's Sacramento–San Joaquin Delta Conflict: From Cooperation to Chicken

Kaveh Madani, Jay R. Lund
2012 Journal of water resources planning and management  
This paper traces changes in this conflict in game-theoretic terms, with its implications for the region's physical and ecological decline and governance.  ...  The State of California may become the victim (or chicken) of the Delta game, bearing the greatest costs, if it continues to rely on leaving parties to develop voluntary cooperative solutions without a  ...  When playing repeated games, players change their strategies over time and adopt a mixture of strategies on the basis of learning about the behavioral characteristics of the other players in the game,  ... 
doi:10.1061/(asce)wr.1943-5452.0000164 fatcat:77m7jenkyjg65ddudqc6djo5ze

Tamer Başar [People in Control]

2021 IEEE Control Systems  
), and how does all this affect equilibria?  ...  These all involve the dependence of equilibria on the kind of information players acquire during the decision-making process, who communicates with whom, and how actions of a particular player affect the  ... 
doi:10.1109/mcs.2021.3107761 fatcat:hivhw75dtvfkneirsqqhpluw4a

Innovation and Strategic Network Formation [article]

Krishna Dasaratha
2022 arXiv   pre-print
Their decisions determine interaction rates between firms, and these interaction rates enter our model as link probabilities in a learning network.  ...  We study a model of innovation with a large number of firms that create new technologies by combining several discrete ideas. These ideas are created via private investment and spread between firms.  ...  We will show that there exist levels of openness q such that for all i the choice q i is optimal given p and q −i .  ... 
arXiv:1911.06872v4 fatcat:f2czdknbufcgjms2kqhug2qr3e

Optimality and Stability in Federated Learning: A Game-theoretic Approach [article]

Kate Donahue, Jon Kleinberg
2021 arXiv   pre-print
One branch of this research has taken a game-theoretic approach, and in particular, prior work has viewed federated learning as a hedonic game, where error-minimizing players arrange themselves into federating  ...  In this work, we motivate and define a notion of optimality given by the average error rates among federating agents (players).  ...  Avrim Blum, Nika Haghtalab, Richard Lanas Phillips, and Han Shao. One for one, or all for all: Equilibria and optimality of collaboration in federated learning. arXiv preprint arXiv:2103.03228, 2021.  ... 
arXiv:2106.09580v1 fatcat:fcnrs62tbfbivo6jceyzdby66m

Competing for congestible goods: experimental evidence on parking choice

María Pereda, Juan Ozaita, Ioannis Stavrakakis, Angel Sánchez
2020 Scientific Reports  
Our results give insights on how to deal with parking problems such as the design of parking lots in central locations in cities and open the way to better understand similar congestible goods problems  ...  We study experimentally a congestible goods problem of relevance for parking design, namely how people choose between a convenient parking lot with few spots and a less convenient one with unlimited space  ...  Clearly, the optimal decision requires coordination and collaboration among the players and is not feasible in the uncoordinated environment we consider here.  ... 
doi:10.1038/s41598-020-77711-w pmid:33257701 fatcat:pmregw6l7zf27a3axztzareh5m

Data Pricing in Machine Learning Pipelines [article]

Zicun Cong, Xuan Luo, Pei Jian, Feida Zhu, Yong Zhang
2021 arXiv   pre-print
We also investigate pricing in the step of collaborative training of machine learning models, and overview pricing machine learning models for end users in the step of machine learning deployment.  ...  Then, we focus on pricing in three important steps in machine learning pipelines. To understand pricing in the step of training data collection, we review pricing raw data sets and data labels.  ...  To succeed in building a machine learning application, one party is far from enough. Many parties have to collaborate in one way or another.  ... 
arXiv:2108.07915v1 fatcat:736zip2pbndupl7hixdbfz33om

Privacy-Preserving Federated Learning for UAV-Enabled Networks: Learning-Based Joint Scheduling and Resource Management [article]

Helin Yang, Jun Zhao, Zehui Xiong, Kwok-Yan Lam, Sumei Sun, Liang Xiao
2020 arXiv   pre-print
However, due to the privacy concerns of devices and limited computation or communication resource of UAVs, it is impractical to send raw data of devices to UAV servers for model training.  ...  In this paper, we develop an asynchronous federated learning (AFL) framework for multi-UAV-enabled networks, which can provide asynchronous distributed computing by enabling model training locally without  ...  [19] proposed an FL-based sensing and collaborative learning approach for UAV-enabled internet of vehicles (IoVs), where UAVs collect data and train AI model for IoVs. In addition, Ng et al.  ... 
arXiv:2011.14197v1 fatcat:jrwqdrdfkbh3joqa2nichlclxu
« Previous Showing results 1 — 15 out of 1,697 results