The goal of data selection is to capture the most structural information from a set of data. This paper presents a fast and accurate data selection method, in which the selected samples are optimized to span the subsp...
详细信息
ISBN:
(纸本)9781728132945
The goal of data selection is to capture the most structural information from a set of data. This paper presents a fast and accurate data selection method, in which the selected samples are optimized to span the subspace of all data. We propose a new selection algorithm, referred to as iterative projection and matching (IPM), with linear complexity w.r.t. the number of data, and without any parameter to be tuned. In our algorithm, at each iteration, the maximum information from the structure of the data is captured by one selected sample, and the captured information is neglected in the next iterations by projection on the null-space of previously selected samples. The computational efficiency and the selection accuracy of our proposed algorithm outperform those of the conventional methods. Furthermore, the superiority of the proposed algorithm is shown on active learning for video action recognition dataset on UCF-101;learning using representatives on Im-ageNet;training a generative adversarial network (GAN) to generate multi-view images from a single-view input on CMU Multi-PIE dataset;and video summarization on UTE Egocentric dataset.
We study how to invest optimally in a stock market having a finite number of assets from a signal processing perspective. In particular, we introduce a portfolio selection algorithm that maximizes the expected cumulat...
详细信息
ISBN:
(纸本)9781479903573
We study how to invest optimally in a stock market having a finite number of assets from a signal processing perspective. In particular, we introduce a portfolio selection algorithm that maximizes the expected cumulative wealth in i.i.d. two-asset discrete-time markets where the market levies proportional transaction costs in buying and selling stocks. This is achieved by using "threshold rebalanced portfolios", where trading occurs only if the portfolio breaches certain thresholds. Under the assumption that the relative price sequences have log-normal distribution from the Black-Scholes model, we evaluate the expected wealth under proportional transaction costs and find the threshold rebalanced portfolio that achieves the maximal expected cumulative wealth over any investment period.
In this paper, a risk-aware robotic sensor network (RSN) is proposed in the context of Critical Infrastructure Protection. Such a network will be comprised of mobile sensor nodes that perceive various aspects of their...
详细信息
ISBN:
(纸本)9781467347020
In this paper, a risk-aware robotic sensor network (RSN) is proposed in the context of Critical Infrastructure Protection. Such a network will be comprised of mobile sensor nodes that perceive various aspects of their environment and topologically reconfigure in order to secure a strategic area of interest. Risk awareness is provided through the application of a recently developed Risk Management Framework to the RSN. The risk level of each node is assessed in terms of their degree of distress, proximity factor, and terrain maneuverability. Risk monitoring alerts are issued whenever any given sensor node's quantitative risk metric exceeds a user-defined threshold value. At this point, a node-in-distress (NID) has been identified as the weak point of the securing structure around which the RSN is deployed. The NID can no longer be used with confidence and the effective perimeter coverage of the RSN has been reduced, thus creating potential security breaches in the area of interest. In response, the remaining nodes will self-organize to maximize the perimeter coverage while minimizing the cost of doing so. A limited set of contingency network topologies is produced via evolutionary multi-objective optimization using the Non-Dominated Sorting Genetic algorithm (NSGA-II) and then ranked according to a human-guided alternative selection algorithm. The security operator picks the most suitable topology, which is then effectuated upon the environment. Results indicate that NSGA-II is capable of producing feasible network topologies to satisfy maximum perimeter coverage, while reducing the energy required for topology reconfiguration. As far as we are concerned, this is the first time a RSN applied to a CIP scenario is self-organized in response to a risk analysis conducted on every sensor node on the basis of multiple risk features.
As a consequence of the diffusion of wireless systems operating in the 2.4 GHz ISM band, harmful interference among heterogeneous networks is becoming a serious issue for their performance. This paper is focused on IE...
详细信息
ISBN:
(纸本)9781424425167
As a consequence of the diffusion of wireless systems operating in the 2.4 GHz ISM band, harmful interference among heterogeneous networks is becoming a serious issue for their performance. This paper is focused on IEEE 802.15.4 WSNs undergoing the interference of co-located IEEE 802.11b/g WLANs or Bluetooth piconets. On the basis of energy measurements carried out using a standard-compliant testbed, the statistics of the interfering energy are characterized for an extensive set of traffic conditions. Then the impact of each interference pattern onto a practical WSN application is analyzed in terms of packet loss rate at the application layer. The last part of the paper introduces a channel selection algorithm based on the estimation of the information-theoretic capacity of the considered channels.
As the number of deployed sensor networks globally increased to billions, data collection from these networks became quite a challenging task. The wide proliferation of smartphones provides an inexpensive solution for...
详细信息
ISBN:
(数字)9781728125619
ISBN:
(纸本)9781728125626
As the number of deployed sensor networks globally increased to billions, data collection from these networks became quite a challenging task. The wide proliferation of smartphones provides an inexpensive solution for the data collection challenge through opportunistic connections with sensors. However, the varying and unreliable nature of opportunistic data collection may lead to significant waste in the limited resources of the sensors. Thus, efficient management of the resources of the sensors is critical to ensure balance between conserving the energy of the sensors and minimizing the redundancy in data transfer from the sensor to the smartphone to the cloud through the smartphones Internet connection. Accordingly, this paper proposes solutions for resource management in opportunistic sensor networks based on a novel sensor selection algorithm and fuzzy logic. The sensor selection algorithm is used to limit the redundancy in data collected from sensors within close proximity, thus saving their energy. On the other hand, fuzzy logic is used to predict the presence of opportunistic connections based on the history of the sensor. Performance evaluation through computer simulations show that these solutions combined lead to significant increase in network lifetime and minimization of redundancy in data collection.
In unstructured P2P content distribution systems, the most Important algorithms to ensure optimal flow of content along multiple dynamically created distribution trees are piece selection algorithms and load balancing...
详细信息
ISBN:
(纸本)9781424429257
In unstructured P2P content distribution systems, the most Important algorithms to ensure optimal flow of content along multiple dynamically created distribution trees are piece selection algorithms and load balancing algorithms. This paper models practical load balancing algorithms and derives a number of insights.
暂无评论