3,350 Hits in 5.5 sec

Real-time Federated Evolutionary Neural Architecture Search [article]

Hangyu Zhu, Yaochu Jin
2020 arXiv   pre-print
This way, we effectively reduce computational and communication costs required for evolutionary optimization and avoid big performance fluctuations of the local models, making the proposed framework well  ...  To address the above challenges, we propose an evolutionary approach to real-time federated neural architecture search that not only optimize the model performance but also reduces the local payload.  ...  Online and Offline Evolutionary NAS As previously discussed, existing evolutionary NAS methods, also including most RL and gradient based NAS algorithms, are meant for offline model optimization.  ... 
arXiv:2003.02793v1 fatcat:cd5jo3xe45gvtflvqrxqtemrhq

Knowledge discovery for scheduling in computational grids

Alexander Fölling, Joachim Lepping
2012 Wiley Interdisciplinary Reviews Data Mining and Knowledge Discovery  
Depending on both the job properties and the current state of the resources those decisions  ...  Acknowledgment The authors would like to thank Frank Hoffmann for his valuable comments and corrections.  ...  Evolutionary Algorithms Evolutionary algorithms are general purpose optimization algorithms that mimic the natural process of the Darwinian evolution and are most successfully applied to nonlinear, global  ... 
doi:10.1002/widm.1060 fatcat:5xtrmrde7zathbyrf76bzgqfoa

Multi-robot path planning using co-evolutionary genetic programming

Rahul Kala
2012 Expert systems with applications  
The other feature of the algorithm includes local optimization using memory based lookup where optimal paths between various crosses in map are stored and regularly updated.  ...  Each robot uses a Grammar based Genetic Programming for figuring the optimal path in a maze-like map, while a master evolutionary algorithm caters to the needs of overall path optimality.  ...  The coarser evolutionary algorithm mainly looked at the static obstacles, and their optimality, the finer evolutionary algorithm tried to escape from all dynamic obstacles using a space-time graph map.  ... 
doi:10.1016/j.eswa.2011.09.090 fatcat:h2lzqkce75agvbvsjzbx3gluvi

From federated learning to federated neural architecture search: a survey

Hangyu Zhu, Haoyu Zhang, Yaochu Jin
2021 Complex & Intelligent Systems  
Then neural architecture search approaches based on reinforcement learning, evolutionary algorithms and gradient-based are presented.  ...  This is followed by a description of federated neural architecture search that has recently been proposed, which is categorized into online and offline implementations, and single- and multi-objective  ...  For example, the method proposed in [12] is a typical offline federated NAS framework using a multi-objective evolutionary algorithm.  ... 
doi:10.1007/s40747-020-00247-z fatcat:3iryqimk6bgkvnnl2ar3x5sdzq

From Federated Learning to Federated Neural Architecture Search: A Survey [article]

Hangyu Zhu, Haoyu Zhang, Yaochu Jin
2020 arXiv   pre-print
Then, neural architecture search approaches based on reinforcement learning, evolutionary algorithms and gradient-based are presented.  ...  This is followed by a description of federated neural architecture search that has recently been proposed, which is categorized into online and offline implementations, and single- and multi-objective  ...  For example, the method proposed in [11] is a typical offline federated NAS framework using a multi-objective evolutionary algorithm.  ... 
arXiv:2009.05868v1 fatcat:tvlftvamajh3fi4rag7u27vyve


A. Gholami, P. Pahlavani, S. Azimi, S. Shakibi
2019 The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences  
The research tries to do indoor positioning using local Wi-Fi fingerprints and signals. To reduce the error to collect local fingerprints, RSS values are recorded in 4 directions and two times.  ...  In this research, a genetic algorithm is used to select the appropriate parameters. Ultimately, the accuracy of the model has reached 1.76 cm.  ...  evolutionary ideas of natural selection and belong to the larger part of evolutionary algorithms (EA).  ... 
doi:10.5194/isprs-archives-xlii-4-w18-441-2019 fatcat:5au7xfum5vebnaejcrgbd37fyy

Open Issues in Evolutionary Robotics

Fernando Silva, Miguel Duarte, Luís Correia, Sancho Moura Oliveira, Anders Lyhne Christensen
2016 Evolutionary Computation  
mapping in the evolution of controllers for complex tasks.  ...  In this article, we review and discuss the open issues in evolutionary robotics.  ...  The authors thank the anonymous reviewers for their constructive feedback and valuable comments.  ... 
doi:10.1162/evco_a_00172 pmid:26581015 fatcat:5bxkcwgfovdtlcvh4rgwqjbu7m


Rahul Kala
2013 Applied Artificial Intelligence  
The MNHS contributes towards optimality of the solution, while the GA gives it an iterative nature and enables the approach to be used on high resolution maps.  ...  The paper focuses upon the use of hybrid Multi-Neuron Heuristic Search (MNHS) and Genetic Algorithm (GA).  ...  This algorithm may consist of two phases: offline phase and online phase. The offline phase deals with the learning of the provided map.  ... 
doi:10.1080/08839514.2013.768880 fatcat:6on6v5tw2jenhhojijvxruhz7i

Bio-inspired Ant Algorithms: A review

Sangita Roy, Sheli Sinha Chaudhuri
2013 International Journal of Modern Education and Computer Science  
In this paper, at first ant algorithms are described in details, then transforms to computational optimization techniques: the ACO metaheuristics and developed ACO algorithms.  ...  The focus of ant algorithms is to find approximate optimized problem solutions using artificial ants and their indirect decentralized communications using synthetic pheromones.  ...  aim is to extract and map boundary"s within the image .  ... 
doi:10.5815/ijmecs.2013.04.04 fatcat:j3z57xy5fffodbwyb6kerrmvjy

Plan Better Amid Conservatism: Offline Multi-Agent Reinforcement Learning with Actor Rectification [article]

Ling Pan, Longbo Huang, Tengyu Ma, Huazhe Xu
2022 arXiv   pre-print
Following this intuition, we propose a simple yet effective method, Offline Multi-Agent RL with Actor Rectification (OMAR), which combines the first-order policy gradients and zeroth-order optimization  ...  Given the recent success of transferring online RL algorithms to the multi-agent setting, one may expect that offline RL algorithms will also transfer to multi-agent settings directly.  ...  We compare OMAR against state-of-the-art offline RL methods including CQL and TD3+BC [Fujimoto and Gu, 2021] . We also compare it with a recent offline MARL algorithm MA-ICQ [Yang et al., 2021] .  ... 
arXiv:2111.11188v3 fatcat:7jza74xvo5bp3pspl2onykwzxm

Cloud Computing Resource Scheduling and a Survey of Its Evolutionary Approaches

Zhi-Hui Zhan, Xiao-Fang Liu, Yue-Jiao Gong, Jun Zhang, Henry Shu-Hung Chung, Yun Li
2015 ACM Computing Surveys  
Because cloud provides a finite pool of virtualized on-demand resources, optimally scheduling them has become an essential and rewarding topic, where a trend of using Evolutionary Computation (EC) algorithms  ...  Cloud computing resource scheduling and a survey of its evolutionary approaches. ACM Comput.  ...  ACKNOWLEDGMENTS The authors would like to thank Yong Wee Foo, Hadoop Administrator and Manager of Communications and Networks Group, Nanyang Polytechnic, Singapore, and University of Glasgow Singapore,  ... 
doi:10.1145/2788397 fatcat:n7bpc276jjgsjnqmiukox7nrqy

Decentralized Innovation Marking for Neural Controllers in Embodied Evolution

Iñaki Fernández Pérez, Amine Boumaza, François Charpillet
2015 Proceedings of the 2015 on Genetic and Evolutionary Computation Conference - GECCO '15  
We compare our method to odNEAT, an algorithm in which agents use local time clocks as innovation numbers, on two multi-robot learning tasks: navigation and item collection.  ...  This method does not rely on a centralized clock, which makes it well suited for the decentralized nature of EE where no central evolutionary process governs the adaptation of a team of robots exchanging  ...  The algorithm is a centralized offline EA with one global population initialized with minimal fully-connected perceptrons, and where all new genes can be tracked since all information is centralized.  ... 
doi:10.1145/2739480.2754759 dblp:conf/gecco/PerezBC15 fatcat:axwk2seoxraxbpdszy73hcbdsa

Data-Driven Offline Optimization For Architecting Hardware Accelerators [article]

Aviral Kumar, Amir Yazdanbakhsh, Milad Hashemi, Kevin Swersky, Sergey Levine
2022 arXiv   pre-print
In this paper, we develop such a data-driven offline optimization method for designing hardware accelerators, dubbed PRIME, that enjoys all of these properties.  ...  optimization.  ...  ACKNOWLEDGEMENTS We thank the "Learn to Design Accelerators" team at Google Research and the Google EdgeTPU team for their invaluable feedback and suggestions.  ... 
arXiv:2110.11346v3 fatcat:ldkyj4enqbh5posziuozl5tj74

Search-Based Procedural Content Generation [chapter]

Julian Togelius, Georgios N. Yannakakis, Kenneth O. Stanley, Cameron Browne
2010 Lecture Notes in Computer Science  
Recently, a small number of papers have appeared in which the authors implement stochastic search algorithms, such as evolutionary computation, to generate game content, such as levels, rules and weapons  ...  The relation between search-based and other types of procedural content generation is described, as are some of the main research challenges in this new field.  ...  The research was supported in part by the Danish Research Agency, Ministry of Science, Technology and Innovation; project name: AGameComIn; project number: 274-09-0083.  ... 
doi:10.1007/978-3-642-12239-2_15 fatcat:xzqvtxiavva6nmk7yptjmqdmuy

Advances in applying soft computing techniques for big data and cloud computing

B. B. Gupta, Dharma P. Agrawal, Shingo Yamaguchi, Michael Sheng
2018 Soft Computing - A Fusion of Foundations, Methodologies and Applications  
This special issue presents some selected papers dealing with different aspects of big data and cloud computing research issues and other related areas and also emphasizes many open questions (Hossain  ...  In addition, we are also grateful to all the authors for submitting and improving their papers.  ...  It is shown that the compressed radio maps based on SKPCA have much smaller sizes than their original radio maps, but achieve similar localization performance and significantly outperform the other two  ... 
doi:10.1007/s00500-018-3575-1 fatcat:lni3r2tqsvdxbp2v6sb33xqa5q
« Previous Showing results 1 — 15 out of 3,350 results