Filters








61,892 Hits in 7.6 sec

Near-Optimal Algorithms for Online Matrix Prediction

Elad Hazan, Satyen Kale, Shai Shalev-Shwartz
2017 SIAM journal on computing (Print)  
By analyzing the decomposability of cut matrices, triangular matrices, and low tracenorm matrices, we derive near optimal regret bounds for online max-cut, online gambling, and online collaborative filtering  ...  In this paper we isolate a property of matrices, which we call (β, τ )decomposability, and derive an efficient online learning algorithm, that enjoys a regret bound of O( √ β τ T ) for all problems in  ...  The Algorithm for Online Matrix Prediction In this section we prove Theorem 1 by constructing an efficient algorithm for Online Matrix Prediction and analyze its regret.  ... 
doi:10.1137/120895731 fatcat:4ba2otdhrrcx5piezdezuh5wh4

Near-Optimal Algorithms for Online Matrix Prediction [article]

Elad Hazan, Satyen Kale, Shai Shalev-Shwartz
2012 arXiv   pre-print
By analyzing the decomposability of cut matrices, triangular matrices, and low trace-norm matrices, we derive near optimal regret bounds for online max-cut, online gambling, and online collaborative filtering  ...  In this paper we isolate a property of matrices, which we call (beta,tau)-decomposability, and derive an efficient online learning algorithm, that enjoys a regret bound of O*(sqrt(beta tau T)) for all  ...  The Algorithm for Online Matrix Prediction In this section we prove Theorem 1 by constructing an efficient algorithm for Online Matrix Prediction and analyze its regret.  ... 
arXiv:1204.0136v1 fatcat:6jhsc2asvfharetxem3zlkrxly

Online Optimization of Collaborative Web Service QoS Prediction Based on Approximate Dynamic Programming

Xiong Luo, Hao Luo, Xiaohui Chang
2015 International Journal of Distributed Sensor Networks  
Therefore, the near-optimal performance of QoS prediction can be achieved. Experimental studies are carried out to demonstrate the effectiveness of the proposed ADP-based prediction approach.  ...  Unlike the traditional QoS prediction approaches, our algorithm in this paper is realized by incorporating approximate dynamic programming-(ADP-) based online parameter tuning strategy into the QoS prediction  ...  We propose a model-free online ADP learning algorithm for parameter tuning used in collaborative web service QoS prediction.  ... 
doi:10.1155/2015/452492 fatcat:xsbjpk6jzrbqdbhr76bdxyskze

OFS Technique with DE Algorithm based on Correlation and Clustering method along with its Application

P R Vhansure, A A Phatak, A S Shimpi
2017 Proceedings of the Second International Conference on Research in Intelligent and Computing in Engineering  
Different from batch learning technique online learning has selected by a motivational scalable, well-organized machine learning algorithm which has been used for large-scale dataset.  ...  For every occurrence the online learning technique should be retrieve complete features/ attributes from large scale dataset volume.  ...  Correlation finds the relation between the dependent variable after that nearest neighboring algorithm used which finds near value for creating the cluster depends on clusters DE algorithm optimize the  ... 
doi:10.15439/2017r8 dblp:conf/rice/VhansurePS17 fatcat:3nnwh4zfrvgizm25rc3eshectm

Structure Parameter Optimized Kernel Based Online Prediction with a Generalized Optimization Strategy for Nonstationary Time Series [article]

Jinhua Guo, Hao Chen, Jingxin Zhang, Sheng Chen
2021 arXiv   pre-print
In this paper, sparsification techniques aided online prediction algorithms in a reproducing kernel Hilbert space are studied for nonstationary time series.  ...  For structure parameters, the kernel dictionary is selected by some sparsification techniques with online selective modeling criteria, and moreover the kernel covariance matrix is intermittently optimized  ...  The online tunable RBF algorithms can obtain better prediction performances than the KAF algorithms in the first cascade-connected group, but this is not the case for the optimal prediction performances  ... 
arXiv:2108.08180v1 fatcat:bdqce2kihbd7do2u3cmarpmjvi

Online adaptive reinitialization of the constant modulus algorithm

S. Evans, Lang Tong
2000 IEEE Transactions on Communications  
An adaptive reinitialization algorithm for the constant modulus algorithm is proposed that relies on the similarities between the constant modulus and the Wiener equalizer and expands the capabilities  ...  of a previous algorithm.  ...  The algorithm seeks to determine the relative delay of the optimal Wiener equalizer near which the optimal CMA equalizer is presumed to reside.  ... 
doi:10.1109/26.843118 fatcat:75n5lzjjtrbwdkwb3cobm5mf3i

Special issue: Optimization models and algorithms for data science

Panos Parpas, Daniel Ralph, Wolfram Wiesemann
2017 Mathematical programming  
convergence analysis for various existing algorithms for constrained convex optimization.  ...  In the first paper (Max-Norm Optimization for Robust Matrix Recovery), Ethan X.  ...  In the third paper (Near-Optimal Stochastic Approximation for Online Principal Component Estimation), Chris Junchi Li, Mengdi Wang, Han Liu and Tong Zhang study the challenging problem of online algorithms  ... 
doi:10.1007/s10107-017-1217-5 fatcat:nhc36v3vujdslf6cmaakqndg4q

Design of the Nonlinear System Predictor Driven by the Bayesian-Gaussian Neural Network of Sliding Window Data

Yijian Liu, Yanjun Fang
2009 Computer and Information Science  
and sliding window data to realize the online output prediction for the nonlinear dynamic system.  ...  The simulation experiment indicates that the Bayesian-Gaussian NN based on the sliding window data can fulfill the demands of the online identification and prediction of the adaptive nonlinear system.  ...  So the self-adjusted method has deficiencies for the online prediction application of the nonlinear dynamic system.  ... 
doi:10.5539/cis.v2n2p26 fatcat:33cjzqfeorfjpgnvzjz5atamne

Online-updating regularized kernel matrix factorization models for large-scale recommender systems

Steffen Rendle, Lars Schmidt-Thieme
2008 Proceedings of the 2008 ACM conference on Recommender systems - RecSys '08  
Regularized matrix factorization models are known to generate high quality rating predictions for recommender systems.  ...  We propose a generic method for learning RKMF models. From this method we derive an online-update algorithm for RKMF models that allows to solve the new-user/ new-item problem.  ...  For your inquiries please contact info@mymediaproject.org.  ... 
doi:10.1145/1454008.1454047 dblp:conf/recsys/RendleS08 fatcat:yx5wkkyofbf4vff5up7tgcyzpq

Online Data-Enabled Predictive Control [article]

Stefanos Baros, Chin-Yao Chang, Gabriel E. Colon-Reyes, Andrey Bernstein
2020 arXiv   pre-print
We develop an online data-enabled predictive (ODeePC) control method for optimal control of unknown systems, building on the recently proposed DeePC [1].  ...  The proposed ODeePC conceptual-wise resembles standard adaptive system identification and model predictive control (MPC), but it provides a new alternative for the standard methods.  ...  We note that in Algorithm 2, the matrix H is not updated over time as the data is recorded offline and used online to solve the receding horizon predictive control problem.  ... 
arXiv:2003.03866v3 fatcat:vca2ok77endgbgzlhh7xq3ciya

A Novel Machine Learning Aided Antenna Selection Scheme for MIMO Internet of Things

Wannian An, Peichang Zhang, Jiajun Xu, Huancong Luo, Lei Huang, Shida Zhong
2020 Sensors  
The corresponding simulation results verified that the proposed MLCNN-aided AS scheme may be capable of achieving near-optimal capacity performance in real time, and the performance is relatively insensitive  ...  Additionally, applying multi-label concept may significantly improve the prediction accuracy of the trained MLCNN model under correlated large-scale MIMO channel conditions with less training data.  ...  Therefore, we propose a new MLCNN-aided CBAS algorithm for the sake of reducing the online AS processing time.  ... 
doi:10.3390/s20082250 pmid:32316141 fatcat:lod3mgi6vjapxcjevjuxgnjm5m

Thinking fast and slow: Optimization decomposition across timescales

Gautam Goel, Niangjun Chen, Adam Wierman
2017 2017 IEEE 56th Annual Conference on Decision and Control (CDC)  
Our results highlight that decomposition of a multi-timescale controller into a fast timescale, reactive controller and a slow timescale, predictive controller can be near-optimal in a strong sense.  ...  In particular, we exhibit such a design, named Multi-timescale Reflexive Predictive Control (MRPC), which maintains a pertimestep cost within a constant factor of the offline optimal in an adversarial  ...  Importantly, competitive algorithms for online convex optimization algorithms do not exist in general, unless the algorithms are given access to noisy predictions about the future.  ... 
doi:10.1109/cdc.2017.8263834 dblp:conf/cdc/GoelCW17 fatcat:54unxwqucfhhxbjcweiwsvvbna

Thinking Fast and Slow: Optimization Decomposition Across Timescales [article]

Gautam Goel, Niangjun Chen, Adam Wierman
2017 arXiv   pre-print
Our results highlight that decomposition of a multi-timescale controller into a fast timescale, reactive controller and a slow timescale, predictive controller can be near-optimal in a strong sense.  ...  In particular, we exhibit such a design, named Multi-timescale Reflexive Predictive Control (MRPC), which maintains a per-timestep cost within a constant factor of the offline optimal in an adversarial  ...  Importantly, competitive algorithms for online convex optimization algorithms do not exist in general, unless the algorithms are given access to noisy predictions about the future.  ... 
arXiv:1704.07785v2 fatcat:tf5cvt5q3rc7poaskepjwnndfu

Thinking Fast and Slow

Gautam Goel, Niangjun Chen, Adam Wierman
2017 Performance Evaluation Review  
Our results highlight that decomposition of a multi-timescale controller into a fast timescale, reactive controller and a slow timescale, predictive controller can be near-optimal in a strong sense.  ...  In particular, we exhibit such a design, named Multi-timescale Reflexive Predictive Control (MRPC), which maintains a pertimestep cost within a constant factor of the offline optimal in an adversarial  ...  Importantly, competitive algorithms for online convex optimization algorithms do not exist in general, unless the algorithms are given access to noisy predictions about the future.  ... 
doi:10.1145/3152042.3152052 fatcat:gcsh36kmofarnms4i33yffhzu4

GPU-Based Lossless Compression of Aurora Spectral Data using Online DPCM

Jiaojiao Li, Jiaji Wu, Gwanggil Jeon
2019 Remote Sensing  
This paper presents a parallel Compute Unified Device Architecture (CUDA) implementation of the prediction-based online Differential Pulse Code Modulation (DPCM) method for the lossless compression of  ...  In the CUDA implementation, we proposed a decomposition method for the matrix multiplication to avoid redundant data accesses and calculations.  ...  Acknowledgments: We thank all the editors and reviewers for their valuable comments that greatly improved the presentation of this paper.  ... 
doi:10.3390/rs11141635 fatcat:7oed237dqfhdzhy7umg4onctb4
« Previous Showing results 1 — 15 out of 61,892 results