检索结果-内蒙古大学图书馆

IEEE Transactions on Artificial Intelligence 2024年第2期5卷 602-612页

作者： Bolluyt, Elijah D. Comaniciu, Cristina Stevens Institute of Technology Department of Electrical and Computer Engineering HobokenNJ07030 United States

This work introduces novel architecture components and training procedures to create augmented neural networks with the ability to process data bidirectionally via an end-to-end approximate inverse. We develop pseudoinvertible neural network (PsI-NN) layers which function as drop-in replacements for corresponding convolutional and fully connected layers;by using these, existing architectures gain a pseudoinverse function which, with training, approximately reverses the forward function. For cases where learning both a task and its inverse are necessary or desirable, we show that PsI-NN enabled models match or exceed the quality of results generated by systems that use two separate models while drastically reducing system parameter count. We demonstrate this on two tasks: unpaired image translation and semisupervised classification. In both, PsI-NN greatly reduces parameter count without any loss of output quality;in semisupervised image classification, PsI-NN improves classification performance beyond the baseline autoencoder method. Our approach creates pseudoinvertible versions of existing architectures, circumventing the stringent constraints required to create a true inverse while allowing a single neural network to learn a task in two directions simultaneously. © 2023 IEEE.

关键词： Supervised learning

来源：评论

学校读者我要写书评

暂无评论

Transforming the electrical grid: the role of AI in advancing smart, sustainable, and secure energy systems

引用

Energy Informatics 2025年第1期8卷 1-43页

作者： Rajaperumal, T.A. Columbus, C. Christopher School of Electrical Engineering Vellore Institute of Technology Tamil Nadu Chennai India School of Computer Science and Engineering Vellore Institute of Technology Tamil Nadu Chennai India

The evolution of the electrical grid from its early centralized structure to today’s advanced "smart grid" reflects significant technological progress. Early grids, designed for simple power delivery from large plants to consumers, faced challenges in efficiency, reliability, and scalability. Over time, the grid has transformed into a decentralized network driven by innovative technologies, particularly artificial intelligence (AI). AI has become instrumental in enhancing efficiency, security, and resilience by enabling real-time data analysis, predictive maintenance, demand-response optimization, and automated fault detection, thereby improving overall operational efficiency. This paper examines the evolution of the electrical grid, tracing its transition from early limitations to the methodologies adopted in present smart grids for addressing those challenges. Current smart grids leverage AI to optimize energy management, predict faults, and seamlessly integrate electric vehicles (EVs), reducing transmission losses and improving performance. However, these advancements are not without limitations. Present grids remain vulnerable to cyberattacks, necessitating the adoption of more robust methodologies and advanced technologies for future grids. Looking forward, emerging technologies such as Digital Twin (DT) models, the Internet of Energy (IoE), and decentralized grid management are set to redefine grid architectures. These advanced technologies enable real-time simulations, adaptive control, and enhanced human–machine collaboration, supporting dynamic energy distribution and proactive risk management. Integrating AI with advanced energy storage, renewable resources, and adaptive access control mechanisms will ensure future grids are resilient, sustainable, and responsive to growing energy demands. This study emphasizes AI’s transformative role in addressing the challenges of the early grid, enhancing the capabilities of the present smart grid, and shaping a secure

关键词： Risk management

来源：评论

学校读者我要写书评

暂无评论

Gaussian Switch Sampling: A Second-Order Approach to Active Learning

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2024年第1期5卷 38-50页

作者： Benkert, Ryan Prabhushankar, Mohit Alregib, Ghassan Pacharmi, Armin Corona, Enrique Georgia Institute of Technology School of Electrical and Computer Engineering AtlantaGA30332-0250 United States Ford Motor Company DearbornMI48126 United States

In active learning, acquisition functions define informativeness directly on the representation position within the model manifold. However, for most machine learning models (in particular neural networks) this representation is not fixed due to the training pool fluctuations in between active learning rounds. Therefore, several popular strategies are sensitive to experiment parameters (e.g., architecture) and do not consider model robustness to out-of-distribution settings. To alleviate this issue, we propose a grounded second-order definition of information content and sample importance within the context of active learning. Specifically, we define importance by how often a neural network 'forgets' a sample during training artifacts of second-order representation shifts. We show that our definition produces highly accurate importance scores even when the model representations are constrained by the lack of training data. Motivated by our analysis, we develop the Gaussian switch sampling (GauSS). We show that GauSS is setup agnostic and robust to anomalous distributions with exhaustive experiments on three in-distribution benchmarks, three out-of-distribution benchmarks, and three different architectures. We report an improvement of up to 5% when compared against four popular query strategies. © 2020 IEEE.

关键词： Neural networks

来源：评论

学校读者我要写书评

暂无评论

TCAD-Based Design of SiGe HBT Germanium Profiles via Bayesian Optimization

TCAD-Based Design of SiGe HBT Germanium Profiles via Bayesia...

引用

2024 IEEE BiCMOS and Compound Semiconductor Integrated Circuits and technology Symposium, BCICTS 2024

作者： Heimerl, Justin P. Teng, Jeffrey W. Lee, Harrison P. Nergui, Delgermaa Moody, Jackson P. Cressler, John D. Georgia Institute of Technology School of Electrical and Computer Engineering AtlantaGA30332-0250 United States

ISBN: (纸本)9798331541248

This work investigates the automatic design of silicon-germanium (SiGe) heterojunction bipolar transistor (HBT) germanium profiles utilizing Bayesian optimization in a commercial TCAD engine. The time-intensive nature of device simulation renders standard numerical optimization strategies infeasible, as gradients are expensive to approximate, and candidate objective functions, such as device minimum noise figure (NFmin), maximum transit frequency fT), or breakdown voltage (BVCEO) are often not well-behaved functions of design variables. Bayesian optimization is a strategy that seeks a global optimum of an expensive-to-evaluate function of which little underlying structure is assumed. Here, we model the output of TCAD simulations as such a function and show that Bayesian optimization is capable of exploring the Ge profile design space, providing optimized profiles in as few as 30 simulation iterations. © 2024 IEEE.

关键词： Noise figure

来源：评论

学校读者我要写书评

暂无评论

Rectenna Characterized Under Varying 2D Transmitter Positions and Power Beaming Amplitude Levels at 5.8 GHz

Rectenna Characterized Under Varying 2D Transmitter Position...

引用

2024 IEEE Opportunity Research Scholars Symposium, ORSS 2024

作者： Mcclelion, Emma K. Raykov, Viktor D. Xiao, Hannah Graves, Kaitlyn M. Durgin, Gregory D. Georgia Institute of Technology School of Electrical and Computer Engineering AtlantaGA30332-0250 United States

ISBN: (纸本)9798350390698

This work provides a reference for attained rectenna conversion efficiency under realistic, non-ideal conditions. Two basic 5.8 GHz rectennas, inlcuding one symmetrical and one non-symmetrical patch antenna, were characterized through simulations under varying non-ideal conditions. Power levels varied from -10 to 50 dBm so to characterize the non-linearity of the diode, and the transmitter-to-rectenna alignment angles varied from -180° to 180°. An orthogonal two-source setup was used to simulate potentially increased input power or coverage in real-world instances. The maximum RF-to-DC conversion efficiency was 78.64%. A maximum receiving antenna collection efficiency of 0.2730% was attained with the symmetrical antenna aligned at 0° with respect to the single-transmitter. The antenna collection efficiency of the single-transmitter system was 1.848-times more efficient than the two-transmitter system. However, the conversion efficiency of the two-transmitter system was more robust against misalignment compared to the single-transmitter system. © 2024 IEEE.

关键词： Transmitters

来源：评论

学校读者我要写书评

暂无评论

Using Pulsed-Mode Measurements of SiGe HBTs for Non-Destructive, Improved RF-SOA Estimation

Using Pulsed-Mode Measurements of SiGe HBTs for Non-Destruct...

引用

2024 IEEE BiCMOS and Compound Semiconductor Integrated Circuits and technology Symposium, BCICTS 2024

作者： Sepúlveda-Ramos, Nelson E. Lee, Harrison P. Teng, Jeffrey W. Cressler, John D. Georgia Institute of Technology School of Electrical and Computer Engineering AtlantaGA30332-0250 United States

ISBN: (纸本)9798331541248

We present a new approach for exploring the underlying reliability physics of SiGe HBTs, and define a new, application-relevant safe-operating area (SOA) of devices by implementing pulsed-voltage measurements. Compared to DC measurements, pulsed-mode measurements have the advantage of being non-destructive, and can be used to decouple self-heating and impact ionization to predict hard failure under large RF swings. In the present work, one of the conventional DC SOA criteria is explored as a function of pulse-width in state-of-the-art SiGe HBTs. The collector-current density (JC) doubles at a later voltage under pulsed conditions vs. DC, resulting in an additional SOA margin of 0.8 V at peak-fT quiescent current. Load-pull measurements show that the devices experience hard failure at about the same bias as the new pulsed-mode SOA definition for the shortest measured pulse-width (160 ns). Load lines overlaid with the pulsed-SOA definition demonstrate that the device experiences hard failure when the RF current swing and voltage are outside the pulsed-mode SOA boundary. These results demonstrate pulsed-mode measurements as a new and useful tool wherein pulsed width can be adjusted to improve the correlation between measurements of the SOA and hard failure under application-relevant RF operating conditions. Furthermore, pulsed-mode SOA can be measured using forced voltage between the base and the emitter, often preferred for power amplifiers, which allows designers to explore more aggressive operational bias conditions that provide circuit peRFormance advantages while remaining robust to damage. © 2024 IEEE.

关键词： Light amplifiers

来源：评论

学校读者我要写书评

暂无评论

DeepER-HD: An Error Resilient HyperDimensional Computing Framework with DNN Front-End for Feature Selection 25

DeepER-HD: An Error Resilient HyperDimensional Computing Fra...

引用

25th IEEE Latin American Test Symposium, LATS 2024

作者： Mejri, Mohamed Amarnath, Chandramouli Chatterjee, Abhijit Georgia Institute Of Technology School Of Electrical And Computer Engineering AtlantaGA30332-0250 United States

ISBN: (纸本)9798350365559

Brain-inspired hyperdimensional (HD) computing models mimic cognition through combinatorial bindings of biological neuronal data represented by high-dimensional vectors and related operations. However, the efficacy of HD computing depends strongly on input signal and data features used to realize such bindings. In this paper, we propose a new HD-computing framework based on a co-trainable DNN-based feature extractor pre-processor and a hyperdimensional computing system. When trained with restrictions on the ranges of hypervector elements for resilience to memory access errors, the framework achieves up to 135% accuracy improvement over baseline HD-computing for error-free operation and up to three orders of magnitude improvement in error resilience compared to the state-of-the-art. Results for a range of applications from image classification, face recognition, human activity recognition and medical diagnosis are presented and demonstrate the viability of the proposed ideas. © 2024 IEEE.

关键词： Face recognition

来源：评论

学校读者我要写书评

暂无评论

Low-Rank Gap Filling and Downscaling for SMAP Soil Moisture Datasets

Low-Rank Gap Filling and Downscaling for SMAP Soil Moisture ...

引用

作者： Beale, Kevin Bras, Rafael L. Romberg, Justin Department of Electrical and Computer Engineering Georgia Institute of Technology AtlantaGA United States Departments of Civil and Environmental Engineering and Earth and Atmospheric Sciences Georgia Institute of Technology AtlantaGA United States

Soil moisture is the linchpin of the surface hydrologic cycle, controlling the partitioning of water and energy fluxes at the surface. Without it, vegetation, and hence life on the solid Earth as we know it, would not exist. Understanding ecohydrology is understanding the availability of soil moisture to vegetation. Until recently, measuring soil moisture was difficult, expensive, intrusive, and local. NASA's Soil Moisture Active Passive (SMAP) mission changed that by providing global estimates at reasonable frequencies. Ecohydrology and many other hydrologic applications are best when high spatiotemporal resolution soil moisture datasets are available. The SMAP and SMAP-Sentinel soil moisture products currently possess contrasting spatial and temporal resolutions, but their coincident nature presents an opportunity to learn how to enhance the spatial resolution of SMAP retrievals to obtain a global, high spatiotemporal resolution dataset. However, a challenge in learning from SMAP-Sentinel data is the presence of missing pixels. In this work, we propose a low-rank approach to both gap-fill SMAP-Sentinel and downscale SMAP and evaluate its performance globally on both held-out SMAP-Sentinel data and measurements from SMAPVEX validation datasets. The proposed method outperformed baselines globally on SMAP-Sentinel data but had mixed performance against retrievals from airborne measurements. A procedure for filling in missing pixels in SMAP-Sentinel measurements using the low-rank models was found to outperform alternative interpolation methods. Overall, the results show that the proposed method can recover missing pixels in soil moisture measurements and can be used to compute estimates of high-resolution SMAP-Sentinel retrievals from low-resolution SMAP data. © 2025 The Author(s). Ecohydrology published by John Wiley & Sons Ltd.

关键词： Orthogonal functions

来源：评论

学校读者我要写书评

暂无评论

E-Track: Eye Tracking with Event Camera for Extended Reality (XR) Applications 5

E-Track: Eye Tracking with Event Camera for Extended Reality...

引用

5th IEEE International Conference on Artificial Intelligence Circuits and Systems, AICAS 2023

作者： Li, Nealson Bhat, Ashwin Raychowdhury, Arijit Georgia Institute of Technology School of Electrical and Computer Engineering Atlanta United States

ISBN: (纸本)9798350332674

Eye tracking is an essential functionality to enable extended reality (XR) applications. However, the latency and power constraints of an XR headset are tight. Unlike fix-rate frame-based RGB cameras, the event camera senses brightness changes and generates asynchronous sparse events with high temporal resolution. Although the event camera exhibits suitable characteristics for eye tracking in XR systems, processing an event-based data stream is a challenging task. In this paper, we present an event-based eye-tracking system that extracts pupil features. It is the first system that operates only with an event camera and requires no additional sensing hardware. We first propose an event-to-frame conversion method that encodes the events triggered by eye motion into a 3-channel frame. Secondly, we train a Convolutional Neural Network (CNN) on 24 subjects to classify the events representing the pupil. Finally, we employ a region of interest (RoI) mechanism that tracks pupil location and reduces the amount of CNN inference by 96%. Our eye-tracking pipeline is able to locate the pupil with an error of 3.68 pixels at 160 mW system power. © 2023 IEEE.

关键词： Cameras

来源：评论

学校读者我要写书评

暂无评论

OATT: Outlier Oriented Alternative Testing and Post-Manufacture Tuning of Mixed-Signal/RF Circuits and Systems

OATT: Outlier Oriented Alternative Testing and Post-Manufact...

引用

2023 IEEE International Test Conference, ITC 2023

作者： Komarraju, Suhasini Tammana, Akhil Amarnath, Chandramouli N. Chatterjee, Abhijit School of Electrical and Computer Engineering Georgia Institute of Technology Atlanta United States

ISBN: (纸本)9798350343250

Prevalent specification-based AMS testing techniques require the use of complex test circuits or regressors that are difficult to implement on-chip as well as suffer from coverage loss when devices are under-specified. Complementary defect based testing techniques require the simulation of explosively large defect sets under assumed failure mechanisms. We overcome these limitations in our proposed approach OATT;an Outlier-oriented Alternative Testing and Tuning methodology. OATT maximizes the number and magnitude of the statistical principal components (PCA) of the time-domain DUT test response vectors across diverse manufacturing process corners. This allows construction of a multi-dimensional Gaussian probability density model that characterizes the distribution of DUT responses in the principal components domain. Outliers of this probability density model are classified as defective devices using calibrated confidence ellipses, implicitly detecting devices with parametric as well as hard defects. The embedded DUT response is acquired using coherent undersampling and does not require explicit signal reconstruction. Post-manufacture tuning is performed by minimizing the statistical distance of the DUT response in the PCA domain from the nominal Gaussian model using multi-arm bandit reinforcement learning. Simulation results demonstrate the viability and promise of the proposed approach. © 2023 IEEE.

关键词： Timing circuits

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：