Distributed Conditional Gradient Online Learning for IoT Optimization

Zhang, Mingchuan; Quan, Wei; Cheng, Nan; Wu, Qingtao; Zhu, Junlong; Zheng, Ruijuan; Li, Keqin

doi:10.1109/jiot.2019.2919562

Cited by 30 publications

(10 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Furthermore, we propose the main results in this paper. To ensure communication between agents, we cite Assumption 1 in [25]. Assumption 1: We assume the communication between agents as a doubly random matrix B = [b i j ] n × n , each term of the doubly stochastic matrix B satisfies b i j > 0 and b ii > 0 only if i, j ∈ V and (i, j) ∈ E.…”

Section: Assumptions and Resultsmentioning

confidence: 99%

See 2 more Smart Citations

Randomised block‐coordinate Frank‐Wolfe algorithm for distributed online learning over networks

Zheng

et al. 2020

Cognitive Computation and Systems

Self Cite

View full text Add to dashboard Cite

The distributed online algorithms which are based on the Frank-Wolfe method can effectively deal with constrained optimisation problems. However, the calculation of the full (sub)gradient vector in those algorithms leads to a huge computational cost at each iteration. To reduce the computational cost of the algorithms mentioned above, the authors present a distributed online randomised block-coordinate Frank-Wolfe algorithm over networks. Each agent in the networks only needs to calculate a subset of the coordinates of its (sub)gradient vector in this algorithm. Furthermore, they make a detailed theoretical analysis of the regret bound of this algorithm. When all local objective functions satisfy the conditions of strongly convex functions, the authors' algorithm attains the regret bound of O(T), where T is the total number of iterations. Furthermore, the theorem results are verified via simulation experiments, which show that the algorithm is effective and efficient.

show abstract

Section: Assumptions and Resultsmentioning

confidence: 99%

“…Regret is the cumulative error between the loss of the decision chosen in each iteration and the loss of the current best decision. It is defined in [25] as follows:…”

Section: Regretmentioning

confidence: 99%

See 1 more Smart Citation

Randomised block‐coordinate Frank‐Wolfe algorithm for distributed online learning over networks

Zheng

et al. 2020

Cognitive Computation and Systems

Self Cite

View full text Add to dashboard Cite

show abstract

“…The traffic congestion often brings a series of problems such as environmental pollution, which greatly reduces the quality of urban life. To improve the urban living environment, there are emerging fields such as smart cities and Internet of Things (IoTs) [1,2], which mainly utilize various information technologies to optimize urban resources and services. As an important part of smart cities, Intelligent Transportation Systems (ITSs) can effectively alleviate urban traffic congestion [3,4].…”

Section: Introductionmentioning

confidence: 99%

An AutoEncoder and LSTM-Based Traffic Flow Prediction Method

Wei

2019

Sensors

142

View full text Add to dashboard Cite

Smart cities can effectively improve the quality of urban life. Intelligent Transportation System (ITS) is an important part of smart cities. The accurate and real-time prediction of traffic flow plays an important role in ITSs. To improve the prediction accuracy, we propose a novel traffic flow prediction method, called AutoEncoder Long Short-Term Memory (AE-LSTM) prediction method. In our method, the AutoEncoder is used to obtain the internal relationship of traffic flow by extracting the characteristics of upstream and downstream traffic flow data. Moreover, the Long Short-Term Memory (LSTM) network utilizes the acquired characteristic data and the historical data to predict complex linear traffic flow data. The experimental results show that the AE-LSTM method had higher prediction accuracy. Specifically, the Mean Relative Error (MRE) of the AE-LSTM was reduced by 0.01 compared with the previous prediction methods. In addition, AE-LSTM method also had good stability. For different stations and different dates, the prediction error and fluctuation of the AE-LSTM method was small. Furthermore, the average MRE of AE-LSTM prediction results was 0.06 for six different days.

show abstract

“…e authors in [27] presented a distributed online algorithm based on a primal-dual dynamic mirror descent for a problem with time-varying coupling inequality constraints and obtained a dynamic regret bound. e authors in [28] proposed a distributed online conditional gradient algorithm for a constrained distributed online optimization problem in the Internet of ings. e existing distributed online optimization algorithm based on the gradient method is simple to calculate and requires little storage; however, to ensure the convergence of the algorithm, the iterative step length usually needs to decrease with an increase in the number of iterations, which will lead to a zigzag path at the end of the algorithm.…”

Section: Introductionmentioning

confidence: 99%

A Distributed Conjugate Gradient Online Learning Method over Networks

Zhu

Shang

et al. 2020

Complexity

Self Cite

View full text Add to dashboard Cite

In a distributed online optimization problem with a convex constrained set over an undirected multiagent network, the local objective functions are convex and vary over time. Most of the existing methods used to solve this problem are based on the fastest gradient descent method. However, the convergence speed of these methods is decreased with an increase in the number of iterations. To accelerate the convergence speed of the algorithm, we present a distributed online conjugate gradient algorithm, different from a gradient method, in which the search directions are a set of vectors that are conjugated to each other and the step sizes are obtained through an accurate line search. We analyzed the convergence of the algorithm theoretically and obtained a regret bound of OT, where T is the number of iterations. Finally, numerical experiments conducted on a sensor network demonstrate the performance of the proposed algorithm.

show abstract

Distributed Conditional Gradient Online Learning for IoT Optimization

Cited by 30 publications

References 0 publications

Randomised block‐coordinate Frank‐Wolfe algorithm for distributed online learning over networks

Randomised block‐coordinate Frank‐Wolfe algorithm for distributed online learning over networks

An AutoEncoder and LSTM-Based Traffic Flow Prediction Method

A Distributed Conjugate Gradient Online Learning Method over Networks

Contact Info

Product

Resources

About