Credit Card Fraud Detection is one of the vital issues nowadays which needs to be tackled urgently. In today's world, everyone is shifting to an online and cashless world for easiness in the transaction. However, ...
Credit Card Fraud Detection is one of the vital issues nowadays which needs to be tackled urgently. In today's world, everyone is shifting to an online and cashless world for easiness in the transaction. However, a colossal fraud scheme is running on the other side of this easiness. Daily, many people fall into this trap. This research work is a little contribution to solving this issue. This academic study uses data from the real world to find fraudulent transactions using Machine Learning techniques such as Decision Trees, Logistics Regression, and Random Forest. Furthermore, Synthetic Minority Oversampling Technique is employed to solve the dataset's imbalance issue. Following that, the effectiveness of machine learning methods is compared by using the “With SMOTE” and “Without SMOTE” techniques.
Intention prediction (IP) is a challenging task for intelligent vehicle’s perception systems. IP provides the likelihood, or probability, of a target vehicle to perform a maneuver subjected to a finite set of possibi...
Intention prediction (IP) is a challenging task for intelligent vehicle’s perception systems. IP provides the likelihood, or probability, of a target vehicle to perform a maneuver subjected to a finite set of possibilities. There are many factors that influence the decision-making process of a driver, which should be considered in a prediction framework. In addition, the lack of labeled large-scale dataset with maneuver annotation imposes another challenge to the task. This paper proposes an Interaction-aware Maneuver Prediction framework, called IAMP, using interaction graphs to extract complex interaction features from traffic scenes. In addition, we explored a semi-supervised approach called Noisy Student to take advantage of unlabeled data in the training step. Experimental results show relevant improvement when using unlabeled data, increasing the average performance of a classifier by 7.17% of accuracy. Moreover, this approach also made it possible to obtain an intention predictor with similar results to a classifier., even when using a shorter observation horizon.
The rise of Industry 4.0 has revolutionized manufacturing by integrating real-time data analysis, artificial intelligence (AI), automation, and interconnected systems, enabling adaptive and resilient smart factories. ...
详细信息
This paper addresses the problem of detecting humans in RGB and Thermal (long-wave IR) images taken by cameras mounted onboard a mobile robot. Human/Pedestrian detection is currently one of the most pertinent object d...
This paper addresses the problem of detecting humans in RGB and Thermal (long-wave IR) images taken by cameras mounted onboard a mobile robot. Human/Pedestrian detection is currently one of the most pertinent object detection problems, mainly due to safety concerns in autonomous vehicles. The majority of approaches apply deep-learning techniques based solely on RGB images. However, they have a few shortcomings, namely that during foggy weather, nighttime, and low-light scenarios, these images may not contain sufficient information. To address these issues, this work studies the use of thermal cameras as a complementary source of information for human detection in indoor and outdoor environments. The proposed approach uses YOLOv5 to detect pedestrians in both thermal and RGB images. Moreover, the different modalities are combined using early and late fusion techniques. Evaluation of the proposed approach is carried out in the FLIR Aligned dataset and in a new in-house dataset. Results indicate that the use of fusion techniques highlights a promising way to improve the overall performance in this application domain.
作者：
Bi, JianWu, QianliangQian, JianjunLuo, LeiYang, JianPCA Lab
Key Lab of Intelligent Perception and Systems for High-Dimensional Information of Ministry of Education School of Computer Science and Engineering Nanjing University of Science and Technology China
With the rapid advancement of 3D scanning technology, point clouds have become a crucial data type in computer vision and machine learning. However, learning robust representations for point clouds remains a significa...
详细信息
This ork presents an approach to the flexibility of energy consumption in Renewable Energy Communities (RECs). A two-stage model for quantifying the flexibility provided by the domestic energy resources operation and ...
This ork presents an approach to the flexibility of energy consumption in Renewable Energy Communities (RECs). A two-stage model for quantifying the flexibility provided by the domestic energy resources operation and its negotiation in a market platform is proposed. In stage 1, the optimal consumption of each prosumer is determined, as well as the respective technical flexibility of their resources, namely the maximum and minimum resource operation limits. In stage 2, this technical flexibility is offered in a local flexibility-only market structure, in which both the DSO and the prosumers can present their flexibility needs and requirements. The flexibility selling and buying bids of the prosumers participating in the market are priced based on their base tariff, which is the energy cost of the prosumers corresponding to their optimal schedule of the first stage when no flexibility is provided. Therefore, providing flexibility is an incentive to reduce their energy bill or increase their utility, encouraging their participation in the local flexibility market.
Quantum Amplitude Estimation (QAE) is an important quantum algorithm that has the potential to quadratically speed up Monte Carlo based *** this paper, we present a variant of the QAE without Phase Estimation Algorith...
Quantum Amplitude Estimation (QAE) is an important quantum algorithm that has the potential to quadratically speed up Monte Carlo based *** this paper, we present a variant of the QAE without Phase Estimation Algorithm called Iterative Refinement QAE (IRQAE). IRQAE can refine the current estimation to a more accurate estimation iteratively, hence it can provide an estimation with arbitrary required accuracy ∊. The key idea of IRQAE is to use a rotation gate to create a quantum state for samplings with the current estimation. Using this idea, we show that IRQAE can provide a highly accurate estimation with lower classical computational complexity and with the same quantum computational complexity compared to state-of-the-art QAEs without phase estimation using numerical experiments. We prove that the computational complexity of IRQAE of the quantum part is O(1/∊) and the classical one is O(1/∊). The quantum cost gives a quadratic advantage over that of the classical Monte Carlo simulation.
Efficient motion and appearance modeling are critical for vision-based Reinforcement Learning (RL). However, existing methods struggle to reconcile motion and appearance information within the state representations le...
Efficient motion and appearance modeling are critical for vision-based Reinforcement Learning (RL). However, existing methods struggle to reconcile motion and appearance information within the state representations learned from a single observation encoder. To address the problem, we present Synergizing Interactive Motion-appearance Understanding (Simoun), a unified framework for vision-based RL Given consecutive observation frames, Simoun deliberately and interactively learns both motion and appearance features through a dual-path network architecture. The learning process collaborates with a structural interactive module, which explores the latent motion-appearance structures from the two network paths to leverage their complementarity. To promote sample efficiency, we further design a consistency-guided curiosity module to encourage the exploration of under-learned observations. During training, the curiosity module provides intrinsic rewards according to the consistency of environmental temporal dynamics, which are deduced from both motion and appearance network paths. Experiments conducted on Deep-Mind control suite and CARLA automatic driving benchmarks demonstrate the effectiveness of Simoun, where it performs favorably against state-of-the-art methods.
作者：
Katalin M. HangosSystems and Control Laboratory
Institute for Computer Science and Control Hungary and Department of Electrical Engineering and Information Systems University of Pannonia Hungary
Decomposition offers the potential to reduce the complexity of model-based optimization, prediction, control and diagnosis by accounting for the structure and sparsity of the describing model. Motivated by this fact, ...
ISBN:
(纸本)9781450397117
Decomposition offers the potential to reduce the complexity of model-based optimization, prediction, control and diagnosis by accounting for the structure and sparsity of the describing model. Motivated by this fact, a rich and powerful collection of decomposition methods are available for model based diagnosis of large-scale complex dynamic systems, too. At the same time, one usually does not have enough information about a large-scale complex dynamic system to construct its precise enough model, so a kind of qualitative dynamic model is often used for the diagnosis [1]. Two structural decomposition based qualitative diagnostic methods are presented in this lecture, together with their component-driven system decomposition ***, a model-based diagnostic method is described that is able to detect and isolate non-technical losses (illegal loads) in low voltage electrical grids of one transformer area [2]. As a preliminary off-line step of the diagnosis, a powerful electrical decomposition method is proposed, which breaks down the overall network to subsystems with one feeder layout enabling to make the necessary computations efficient. The diagnostic method is based on analyzing the differences between the measured and model-predicted voltages. The uncertainty in the model parameters together with the measurement uncertainties are also taken into account to make the approach applicable in real-world cases. The proposed method is able to detect and localize multiple illegal loads, and the amount of the illegal consumption can also be *** a second case study, a high level decomposition approach for process system fault diagnosis using event traces is given [3], [4]. Using a simple component graph model behind the process system and the measured trace applied for the diagnosis, the method can find the root cause(s) of propagating failures between separate components. The method can connect individually operating lower-level component-specific diag
Two-dimensional echocardiography is the most widely used non-invasive imaging modality due to its fast acquisition time, low cost, and high temporal resolution. Accurate segmentation of the left ventricle in echocardi...
Two-dimensional echocardiography is the most widely used non-invasive imaging modality due to its fast acquisition time, low cost, and high temporal resolution. Accurate segmentation of the left ventricle in echocardiography is vital for ensuring the accuracy of subsequent diagnosis. Currently, numerous efforts have been made to automatize this task and various public datasets have been released in recent decades to further develop present research. However, medical datasets acquired at different institutions have inherent bias caused by various confounding factors, such as operation policies, machine protocols, treatment preference, etc. As a result, models trained on one dataset, regardless of volume, cannot be confidently utilized for the others. In this study, we investigated model robustness to dataset bias using two publicly available echocardiographic datasets. This work validates the efficacy of a supervised deep learning model for left ventricle segmentation and ejection fraction prediction, outside the dataset on which it was trained. The exposure of this model to unseen, but related samples without additional training maintained a good performance. However, a performance decrease from the original results can be observed, while the impact of quality is also noteworthy with lower quality data leading to decreased performance.
暂无评论