Deep neural networks (DNNs) have emerged as the most effective programming paradigm for computer vision and natural language processing applications. With the rapid development of DNNs, efficient hardware architecture...
详细信息
Deep neural networks (DNNs) have emerged as the most effective programming paradigm for computer vision and natural language processing applications. With the rapid development of DNNs, efficient hardware architectures for deploying DNN-based applications on edge devices have been extensively studied. Emerging nonvolatile memories (NVMs), with their better scalability, nonvolatility, and good read performance, are found to be promising candidates for deploying DNNs. However, despite the promise, emerging NVMs often suffer from reliability issues, such as stuck-at faults, which decrease the chip yield/memory lifetime and severely impact the accuracy of DNNs. A stuck-at cell can be read but not reprogrammed, thus, stuck-at faults in NVMs may or may not result in errors depending on the data to be stored. By reducing the number of errors caused by stuck-at faults, the reliability of a DNN-based system can be enhanced. This article proposes CRAFT, i.e., criticality-aware fault-tolerance enhancement techniques to enhance the reliability of NVM-based DNNs in the presence of stuck-at faults. A data block remapping technique is used to reduce the impact of stuck-at faults on DNNs accuracy. Additionally, by performing bit-level criticality analysis on various DNNs, the critical-bit positions in network parameters that can significantly impact the accuracy are identified. Based on this analysis, we propose an encoding method which effectively swaps the critical bit positions with that of noncritical bits when more errors (due to stuck-at faults) are present in the critical bits. Experiments of CRAFT architecture with various DNN models indicate that the robustness of a DNN against stuck-at faults can be enhanced by up to 105 times on the CIFAR-10 dataset and up to 29 times on ImageNet dataset with only a minimal amount of storage overhead, i.e., 1.17%. Being orthogonal, CRAFT can be integrated with existing fault-tolerance schemes to further enhance the robustness of DNNs aga
Local advertising has become a popular form of online advertising, blurring the lines between traditional commercials and editorial content. This type of advertising is often presented in the same format as the surrou...
详细信息
In recent years, in response to the global climate crisis, inverter-based energy resources (IBRs) have been massively integrated into power systems. Their effective operation and control lie on a wide variety of inner...
详细信息
Relative overgeneralization (RO) occurs in cooperative multi-agent learning tasks when agents converge towards a suboptimal joint policy due to overfitting to suboptimal behaviors of other *** methods have been propos...
详细信息
Relative overgeneralization (RO) occurs in cooperative multi-agent learning tasks when agents converge towards a suboptimal joint policy due to overfitting to suboptimal behaviors of other *** methods have been proposed for addressing RO in multi-agent policy gradient (MAPG) methods although these methods produce state-of-the-art *** address this gap, we propose a general, yet simple, framework to enable optimistic updates in MAPG methods that alleviate the RO *** approach involves clipping the advantage to eliminate negative values, thereby facilitating optimistic updates in *** optimism prevents individual agents from quickly converging to a local ***, we provide a formal analysis to show that the proposed method retains optimality at a fixed *** extensive evaluations on a diverse set of tasks including the Multi-agent MuJoCo and Overcooked benchmarks, our method outperforms strong baselines on 13 out of 19 tested tasks and matches the performance on the rest. Copyright 2024 by the author(s)
This paper presents a performance analysis of novel doubledampedtuned alternating current (AC) filters in high voltage direct current(HVDC) systems. The proposed double-damped tuned AC filters offer theadvantages of i...
详细信息
This paper presents a performance analysis of novel doubledampedtuned alternating current (AC) filters in high voltage direct current(HVDC) systems. The proposed double-damped tuned AC filters offer theadvantages of improved performance of HVDC systems in terms of betterpower quality, high power factor, and lower total harmonic distortion (THD).The system under analysis consists of an 878 km long HVDC transmissionline connecting converter stations at Matiari and Lahore, two major cities inPakistan. The main focus of this research is to design a novel AC filter usingthe equivalent impedance method of two single-tuned and double-dampedtuned AC filters. Additionally, the impact of the damping resistor on the ACchannel is examined. TheTHDof theHVDCsystem with and without currentAC filters was also compared in this research and a double-damped tuned ACfilter was proposed. The results of the simulation represent that the proposeddouble-damped tuned AC filter is far smaller in size, offers better powerquality, and has a much lower THD compared to the AC filters currently inplace in the converter station. The simulation analysis was carried out utilizingpower systems computer-aided design (PSCAD) software.
Advancements in unmanned aerial vehicle (UAV) technology, along with indoor hybrid LiFi-WiFi networks (HLWN), promise the development of cost-effective, energy-efficient, adaptable, reliable, rapid, and on-demand HLWN...
详细信息
This study presents the development and implementation of a sophisticated Web Application Firewall (WAF) empowered by machine learning techniques to bolster cybersecurity measures. Traditional WAFs primarily rely on r...
详细信息
Skin cancer has become a major global health issue, necessitating improved diagnostic methods for better clinical outcomes. This study thoroughly evaluates deep learning methods for classifying skin cancer using dermo...
详细信息
We present the first learning-augmented data structure for implementing dictionaries with optimal consistency and *** data structure, named RobustSL, is a skip list augmented by predictions of access frequencies of el...
详细信息
We present the first learning-augmented data structure for implementing dictionaries with optimal consistency and *** data structure, named RobustSL, is a skip list augmented by predictions of access frequencies of elements in a data *** proper predictions, RobustSL has optimal consistency (achieves static optimality).At the same time, it maintains a logarithmic running time for each operation, ensuring optimal robustness, even if predictions are generated ***, RobustSL has all the advantages of the recent learning-augmented data structures of Lin, Luo, and Woodruff (ICML 2022) and Cao et al.(arXiv 2023), while providing robustness guarantees that are absent in the previous *** experiments show that RobustSL outperforms alternative data structures using both synthetic and real datasets. Copyright 2024 by the author(s)
In this paper, we introduce a new class of score-based generative models (SGMs) designed to handle high-cardinality data distributions by leveraging concepts from mean-field theory. We present mean-field chaos diffusi...
详细信息
In this paper, we introduce a new class of score-based generative models (SGMs) designed to handle high-cardinality data distributions by leveraging concepts from mean-field theory. We present mean-field chaos diffusion models (MF-CDMs), which address the curse of dimensionality inherent in high-cardinality data by utilizing the propagation of chaos property of interacting particles. By treating high-cardinality data as a large stochastic system of interacting particles, we develop a novel score-matching method for infinite-dimensional chaotic particle systems and propose an approximation scheme that employs a subdivision strategy for efficient training. Our theoretical and empirical results demonstrate the scalability and effectiveness of MF-CDMs for managing large high-cardinality data structures, such as 3D point clouds. Copyright 2024 by the author(s)
暂无评论