An entropy-based framework for dynamic clustering and coverage problems

Sharma, Puneet

doi:10.1109/allerton.2009.5394887

Cited by 10 publications

(18 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the resulting algorithm (popularly known as Deterministic Annealing (DA)) it is observed that the solution changes significantly only at certain critical values of β = β cr that correspond to the instances of phase transition. At other values of β, the solution does not change much [51]. It has been observed that a geometric law…”

Section: Design Of Self Organizing Networkmentioning

confidence: 93%

Parameterized MDPs and Reinforcement Learning Problems—A Maximum Entropy Principle-Based Framework

Srivastava

Salapaka

2022

IEEE Trans. Cybern.

View full text Add to dashboard Cite

We present a framework to address a class of sequential decision making problems. Our framework features learning the optimal control policy with robustness to noisy data, determining the unknown state and action parameters, and performing sensitivity analysis with respect to problem parameters. We consider two broad categories of sequential decision making problems modeled as infinite horizon Markov Decision Processes (MDPs) with (and without) an absorbing state. The central idea underlying our framework is to quantify exploration in terms of the Shannon Entropy of the trajectories under the MDP and determine the stochastic policy that maximizes it while guaranteeing a low value of the expected cost along a trajectory. This resulting policy enhances the quality of exploration early on in the learning process, and consequently allows faster convergence rates and robust solutions even in the presence of noisy data as demonstrated in our comparisons to popular algorithms such as Q-learning, Double Q-learning and entropy regularized Soft Q-learning. The framework extends to the class of parameterized MDP and RL problems, where states and actions are parameter dependent, and the objective is to determine the optimal parameters along with the corresponding optimal policy. Here, the associated cost function can possibly be non-convex with multiple poor local minima. Simulation results applied to a 5G small cell network problem demonstrate successful determination of communication routes and the small cell locations. We also obtain sensitivity measures to problem parameters and robustness to noisy environment data.

show abstract

Section: Design Of Self Organizing Networkmentioning

confidence: 93%

Parameterized MDPs and Reinforcement Learning Problems—A Maximum Entropy Principle-Based Framework

Srivastava

Salapaka

2022

IEEE Trans. Cybern.

View full text Add to dashboard Cite

show abstract

“…Maintaining line of sight (LOS) [36] is often too restrictive, especially when there are many obstacles in the terrain. Clustering based approaches [37], [39] have been proposed to alleviate this problem. These approaches employ flying aerial platforms (AP) as relay nodes that make the communication possible in the case of harsh environments.…”

Section: System Level Design For Maintaining Communication Connecmentioning

confidence: 99%

“…The problem was formulated as a dynamic clustering problem with AP interdistance and capacity constraints. Reference [39] addressed distributed implementation of clustering algorithms and used a dynamic maximum-entropy approach.…”

Section: System Level Design For Maintaining Communication Connecmentioning

confidence: 99%

Communication network challenges for collaborative vehicles

Chen

Hovareshti

Baras

2011

2011 Conference Record of the Forty Fifth Asilomar Conference on Signals, Systems and Computers (ASILOMAR)

View full text Add to dashboard Cite

Networked systems of autonomous mobile agents have emerged in a variety of applications such as collaborative robotics, unmanned aerial/ground vehicles, mobile sensor networks and disaster relief operations. These agents utilize wireless communications for distributed computing, control and decision-making. Due to their limitations on energy supply, design and implementation of efficient distributed algorithms are crucial for these systems. This paper reviews different design aspects of networks of collaborative vehicles. We survey existing methods of addressing physical (PHY) and medium access control (MAC) layer, routing, and geometric connectivity issues for such networks and show limitations of these designs that address only single layer issues. We argue that a system engineering framework is needed for the design of such systems.

show abstract

“…Attention 0 x , different from other features, x 's value can be greater than 1. And can be calculated by GIS from (c) and (d) When GIS runs enough iterations, from (a) (b) (c)(d)(8) and (9), the probability distribution corresponding to ME can be got, and the parameter of (7). Then all of the other (N-n) type-unknown nodes with the probability from (7) are labeled [14].…”

Section: Detail Of Me Algorithmmentioning

confidence: 99%

“…It makes use of principle of maximum entropy to compute the baseline distributions representing traffic behavior. In [9], a Maximum Entropy Principle framework allowing us to address the inherent trade-off between the resolution of the clusters and the computation cost is proposed, and it provides flexibility to a variety of dynamic specifications.…”

Section: Introductionmentioning

confidence: 99%

A Malicious Node Detection Algorithm Based on Principle of Maximum Entropy in WSNs

Dai

Liu

Guo

et al. 2012

JNW

View full text Add to dashboard Cite

WSN is a distributed network exposed to an open environment, which is vulnerable to malicious nodes. To find out malicious nodes among a WSN with mass sensor nodes, this paper presents a malicious detection method based on maximum entropy model. Given the types of a few sensor nodes, it extracts sensor nodes’ preferences related with the known types of malicious node, establishes the maximum entropy model of all sensor nodes that participate in network activities. Then, according to the study on the type-known sensor nodes’ samples based on principle of maximum entropy, all of the unknown-type sensor nodes are classified, with probabilities of different types. The experiment results show that as long as the preferences of sensor nodes are precise and the number of active sensor nodes is stable, the detection rate of malicious nodes is stabilized over 90%

show abstract

An entropy-based framework for dynamic clustering and coverage problems

Cited by 10 publications

References 33 publications

Parameterized MDPs and Reinforcement Learning Problems—A Maximum Entropy Principle-Based Framework

Parameterized MDPs and Reinforcement Learning Problems—A Maximum Entropy Principle-Based Framework

Communication network challenges for collaborative vehicles

A Malicious Node Detection Algorithm Based on Principle of Maximum Entropy in WSNs

Contact Info

Product

Resources

About