Face verification is a well-known image analysis application and is widely used to recognize individuals in contemporary society. However, most real-world recognition systems ignore the importance of protecting the id...
详细信息
Earthquakes are considered to be one of the deadliest natural phenomena that exist on Earth. The destruction during an earthquake is directly related to the magnitude corresponding to a station during the earthquake. ...
详细信息
ISBN:
(数字)9798331505714
ISBN:
(纸本)9798331505721
Earthquakes are considered to be one of the deadliest natural phenomena that exist on Earth. The destruction during an earthquake is directly related to the magnitude corresponding to a station during the earthquake. The magnitude of an earthquake is calculated from the complete time series recorded by the instrument. In recent years, the early phase of the time series has been used to measure an earthquake’s magnitude, commonly known as an earthquake early warning system. This early detection of magnitude gives sufficient time for saving lives and property damaged by the earthquake. Traditionally, simple linear relations are utilized in the EEWS. In this study, the proposed Point Cloud Magnitude Prediction (CloudMag) model utilized point cloud technology, which converts two-dimensional time series data into multi-dimensional representation in space. This representation is then utilized for earthquake magnitude prediction using a convolutional neural network architecture. This study discusses a case of the deadly Osaka earthquake that occurred on 18 June 2018 at 7:58:35 AM Japan Standard Time. The result shows that the novel deep learning approach for magnitude estimation is superior to traditional and other machine learning models.
The rapid increase in the volume of data increased the size and complexity of the deep learning models. These models are now more resource-intensive and time-consuming for training than ever. This paper presents a qua...
详细信息
ISBN:
(数字)9798350354119
ISBN:
(纸本)9798350354126
The rapid increase in the volume of data increased the size and complexity of the deep learning models. These models are now more resource-intensive and time-consuming for training than ever. This paper presents a quantum transfer learning (QTL) based approach to significantly reduce the number of parameters of the classical models without compromising their performance, sometimes even improving it. Reducing the number of parameters reduces overfitting problems and training time and increases the models' flexibility and speed of response. For illustration, we have selected a surface anomaly detection problem to show that we can replace the resource-intensive and less flexible anomaly detection system (ADS) with a quantum transfer learning-based hybrid model to address the frequent emergence of new anomalies better. We showed that we could reduce the total number of trainable parameters up to 90 % of the initial model without any drop in performance.
Stock market nature is nonlinear, and research on it has become increasingly important in recent years. The influence of numerous factors on stock prices renders stock forecasting an intricate and demanding task. Poss...
详细信息
ISBN:
(数字)9798350364866
ISBN:
(纸本)9798350364873
Stock market nature is nonlinear, and research on it has become increasingly important in recent years. The influence of numerous factors on stock prices renders stock forecasting an intricate and demanding task. Possible stock market price forecast increases people's gains while minimizing their risks. Ma-chine learning and deep learning techniques play an important role in forecasting stock prices in today's scenario, considering technical indicators. The choice of deep learning technique and tweaking its hyper-parameters are crucial components of stock price prediction. Several model architectures are suggested here, taking into account LSTM deep learning techniques and hyper-parameters like the batch size, number of hidden layers and number of epochs. Hyper-parameters such as the number of years data, the number of preceding days considered, and database spitting are examined and discussed here, in addition to hyper-parameters relating to LSTM networks. The model with 11 years of historical data, 60 or 100 previous days consideration, and 70:10:20 database split outperforms as compared to other model architectures suggested in the paper.
The research illustrates use of ML algos in the field of housing price prediction. The models have been analyzed on real datasets downloaded from Kaggle created by Amitabha Chakraborty. We know that the source on the ...
详细信息
This paper serves to show how Q-learning can be used in complement with Maximal Ratio Combining to better IoT communication systems under dynamic fading, high-density networks, and severe interference sources. The ML-...
详细信息
ISBN:
(数字)9798350387315
ISBN:
(纸本)9798350387322
This paper serves to show how Q-learning can be used in complement with Maximal Ratio Combining to better IoT communication systems under dynamic fading, high-density networks, and severe interference sources. The ML-enhanced MRC demonstrated promising improvements in signal strength and quality, allowing for performance that was either above or at least equal to the regular MRC. The ML-enhanced presented the added benefit of dynamic adaptation, modifying its performance to meet real-life needs by adapting to issues such as increased density and interference. These results support that integrating Q-learning into MRC-based frameworks significantly increases the reliability and efficiency of IoT communication protocols. Future investigations will aim to improve ML algorithms more appropriate for devices with few resources and evaluate their potential with state-of-the-art technology to foster future applications across intelligent IoT networks. These approaches have the potential to support more efficient operations and increased connectivity in IoT systems.
In this paper, we propose a hybrid clustering algorithm based on hierarchical clustering in combination with string clustering, similar to that used in clustering genetic sequences, to extract clusters of operational ...
详细信息
It is critical for society that we transform our siloed water management and infrastructure systems into smart, connected, sustainable, and resilient systems. This transformation can help us to address the effects of ...
详细信息
ISBN:
(数字)9780784484302
ISBN:
(纸本)9780784484302
It is critical for society that we transform our siloed water management and infrastructure systems into smart, connected, sustainable, and resilient systems. This transformation can help us to address the effects of increasing extreme climate events, ecosystem demands, rapid global urbanization, and infrastructure deterioration from age and neglect. As water utilities improve their asset management programs, it is imperative that the data-driven decision support systems represent the complexities within water pipeline infrastructure systems. Artificial intelligence (AI) techniques can enable modelers to train mathematical algorithms to learn complex patterns from data and represent the water pipeline infrastructure systems accurately. PIPEiD is a national database platform that uses artificial intelligence (AI) and machine learning techniques to assess the performance and risk of water pipelines to help utilities better assess pipe replacement decisions and allocate funding. PIPEiD (Pipeline Infrastructure Database) will assist water sector utilities to manage water pipeline infrastructure systems more effectively for performance, resiliency, and sustainability. PIPEiD will provide the secure, robust, and centralized web-based database platform to address all three major infrastructure asset management levels: strategic, tactical, and operational for utilities of all sizes (small, medium, and large) across the country. The research team collected field performance data for potable, raw, and reuse water pipelines made from materials reflecting the wide range of pipes currently in the ground throughout the US, including cast and ductile iron, prestressed concrete cylinder pipe, reinforced concrete, steel, thermoplastic, PVC, and asbestos. The researchers worked to collect data distributed across different ecological areas, or cohorts, organized based on the climatic conditions of the 500 water utilities' locations. These cohorts included coastal, arid, Arctic, and mo
Floods remain a persistent threat to human lives and infrastructure. Identifying flood-impacted houses in drone imageries will support the rescue teams to prioritize rescue operations in highly affected areas. This st...
详细信息
ISBN:
(数字)9798331540685
ISBN:
(纸本)9798331540692
Floods remain a persistent threat to human lives and infrastructure. Identifying flood-impacted houses in drone imageries will support the rescue teams to prioritize rescue operations in highly affected areas. This study investigates the performance of various deep-learning detection models for post-flood house detection using drone imagery. Our research aims to identify the most suitable model and optimized parameters for accurately and efficiently identifying flood-affected houses, thereby facilitating timely disaster response and recovery efforts. The paper explores various state-of-the-art architectures and provides a thorough performance evaluation.
As concerns over privacy and trust escalate among individuals and corporations, traditional memory-based trust systems are increasingly seen as inadequate and undesirable. The rise of microservice architecture complic...
详细信息
ISBN:
(数字)9798350386745
ISBN:
(纸本)9798350386752
As concerns over privacy and trust escalate among individuals and corporations, traditional memory-based trust systems are increasingly seen as inadequate and undesirable. The rise of microservice architecture complicates this by making data paths harder to track. Given these concerns, the critical question arises: How can we establish trust or verify an entity’s credibility without collecting private information or relying on comparisons with pre-existing data, while knowing the exact data path of each request? To address this challenge, we introduce a lightweight, scalable, ZCube platform that employs nested Zero Knowledge Proofs (ZKPs), eliminating reliance on memory or evaluations based on past actions. This platform verifies and maintains trust in real-time without centralizing trust in any specific part of the system. It anticipates intentions and rigorously monitors them, ensuring operational integrity while preserving privacy. Distinctive for its ZKP trustless setup incorporating blockchain, each request begins anew and is rigorously monitored bidirectionally during execution. Our tests on the realistic example of a microservice-based distributed system, OpenTelemetry Demo, confirm that our novel approach— combining proof segmentation with plan adherence— exhibits fast response times of less than 200ms on average for most test cases. This ZCube platform surpasses traditional methods in efficiency and security. Our analysis shows comprehensive security and enhanced performance while effectively maintaining bidirectional trustlessness.
暂无评论