检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Sen, Mrinmay Qin, A.K. Gayathri, C. Kishore, Raghu N. Chen, Yen-Wei Raman, Balasubramanian Dept. of Computing Technologies Swinburne University of Technology HawthornVIC Australia Dept. of Artificial Intelligence Indian Institute of Technology Hyderabad Hyderabad India Dept. of Computer Science and Engineering Mahindra University Hyderabad India Dept. of Information Science and Engineering Ritsumeikan University Kyoto Japan Dept. of Computer Science and Engineering Indian Institute of Technology Roorkee Roorkee India

This paper introduces a new stochastic optimization method based on the regularized Fisher information matrix (FIM), named SOFIM, which can efficiently utilize the FIM to approximate the Hessian matrix for finding Newton's gradient update in large-scale stochastic optimization of machine learning models. It can be viewed as a variant of natural gradient descent, where the challenge of storing and calculating the full FIM is addressed through making use of the regularized FIM and directly finding the gradient update direction via Sherman-Morrison matrix inversion. Additionally, like the popular Adam method, SOFIM uses the first moment of the gradient to address the issue of non-stationary objectives across mini-batches due to heterogeneous data. The utilization of the regularized FIM and Sherman-Morrison matrix inversion leads to the improved convergence rate with the same space and time complexities as stochastic gradient descent (SGD) with momentum. The extensive experiments on training deep learning models using several benchmark image classification datasets demonstrate that the proposed SOFIM outperforms SGD with momentum and several state-of-the-art Newton optimization methods in term of the convergence speed for achieving the pre-specified objectives of training and test losses as well as test accuracy. Copyright © 2024, The Authors. All rights reserved.

关键词： Fisher information matrix

来源：评论

学校读者我要写书评

暂无评论

Grep: Performance Enhancement in MultiCore Processors using an Adaptive Graph Prefetcher

Grep: Performance Enhancement in MultiCore Processors using ...

引用

IEEE computer Society Annual Symposium on VLSI

作者： Indranee Kashyap Dipika Deb Nityananda Sarma School of Computer Science and Engineering RV University India Dept. of Computer Science and Engineering Tezpur University India Dept. of Computer Science and Engineering Indian Institute of Technology Guwahati India

Memory latency and off-chip bandwidth have been struggling to keep up with computing performance in modern computer systems. In this regard, prefetching helps in masking the long memory access latency at various cache levels by continuously monitoring an application’s memory access pattern. Upon detecting a pattern, it prefetches cache block ahead of its use. However, complex patterns such as directed or indirected pointer access, linked lists, and so on does not adhere to any specific pattern and hence, makes prefetching impossible. The paper proposes Grep, an adaptive graph based data prefetcher that monitors L1D cache misses and prefetches block in L2 cache. Unlike state-of-the-art prefetchers, Grep does not search for patterns in the miss stream. Rather, it generates a predecessor-successor relationship among the cache misses by constructing an occurrence graph that stores the frequency and sequence of subsequent cache block accesses. Therefore, both regular and irregular patterns in the miss stream can be predicted. Upon an address match in the occurrence graph, Grep prefetches block with a confidence value. Experimentally, it improves prefetch coverage and accuracy by 35.5% and 18.8%, respectively, compared to SPP.

关键词：

来源：评论

学校读者我要写书评

暂无评论

SOFIM: Stochastic Optimization Using Regularized Fisher Information Matrix

SOFIM: Stochastic Optimization Using Regularized Fisher Info...

引用

International Joint Conference on Neural Networks (IJCNN)

作者： Mrinmay Sen A. K. Qin Gayathri C Raghu Kishore N Yen-Wei Chen Balasubramanian Raman Dept. of Computing Technologies Swinburne University of Technology Hawthorn Victoria Australia Dept. of Artificial Intelligence Indian Institute of Technology Hyderabad Hyderabad India Dept. of Computer Science and Engineering Mahindra University Hyderabad India Dept. of Information Science and Engineering Ritsumeikan University Kyoto Japan Dept. of Computer Science and Engineering Indian Institute of Technology Roorkee Roorkee India

ISBN: (数字)9798350359312

ISBN: (纸本)9798350359329

This paper introduces a new stochastic optimization method based on the regularized Fisher information matrix (FIM), named SOFIM, which can efficiently utilize the FIM to approximate the Hessian matrix for finding Newton’s gradient update in large-scale stochastic optimization of machine learning models. It can be viewed as a variant of natural gradient descent, where the challenge of storing and calculating the full FIM is addressed through making use of the regularized FIM and directly finding the gradient update direction via Sherman-Morrison matrix inversion. Additionally, like the popular Adam method, SOFIM uses the first moment of the gradient to address the issue of non-stationary objectives across mini-batches due to heterogeneous data. The utilization of the regularized FIM and Sherman-Morrison matrix inversion leads to the improved convergence rate with the same space and time complexities as stochastic gradient descent (SGD) with momentum. The extensive experiments on training deep learning models using several benchmark image classification datasets demonstrate that the proposed SOFIM outperforms SGD with momentum and several state-of-the-art Newton optimization methods in term of the convergence speed for achieving the pre-specified objectives of training and test losses as well as test accuracy.

关键词： Training Deep learning Accuracy Neural networks Stochastic processes Optimization methods Benchmark testing

来源：评论

学校读者我要写书评

暂无评论

Analysis of Stock Market Values Using an Optimal Probability with KNN

Analysis of Stock Market Values Using an Optimal Probability...

引用

Advancements in Smart, Secure and Intelligent Computing (ASSIC), International Conference on

作者： Yedavalli Venkata Raghava Rao Ajmeera Kiran P Chinnasamy RaviKanth Motupalli K. Sai Madhuri Dilip Kumar CSE AIML AVNIETJNTUH AIML DEPT Department of Computer Science and Engineering MLR Institute of Technology Hyderabad Department of CSE VNR Vignana Jyothi Institute of Enginnering and Technology Hyderabad Dept.of Electrical and Electronics Engineering MLR Institute of Technology Hyderabad India

The current trend for saving an individual's economic trouble amid a pandemic is stock prediction. Most businesses and middle-class earners invest in the stock market, but without accurate forecast, investors may lose money if they buy in the wrong company. Many trading apps address this issue by showing firm profits and losses. These apps use typical machine learning yet have inefficient performance measures. The proposed model used optimized probabilistic KNN method to fix the distance metric issues in pure KNN. Most test data in present KNN concentrates on centric points, but in stock prediction, all qualities are connected, hence centric points have little effect. Instead of focal points, the suggested algorithm optimizes KNN distance probabilities to focus on pro data point values. The current KNN model must calculate the distance between every data point and test point, which uses more memory and resources. The technology eliminates comparable records to avoid verification. This cuts iterations and speeds up the process. Comparing the model against benchmark classifiers shows its efficiency.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Minimizing Queue Length Regret for Arbitrarily Varying Channels

arXiv

引用

arXiv 2025年

作者： Krishnakumar, G. Sinha, Abhishek Dept. of Electrical Engineering Indian Institute of Technology Madras Chennai600036 India School of Technology and Computer Science Tata Institute of Fundamental Research Mumbai400005 India

We consider an online channel scheduling problem for a single transmitter-receiver pair equipped with N arbitrarily varying wireless channels. The transmission rates of the channels might be non-stationary and could be controlled by an oblivious adversary. At every slot, incoming data arrives at an infinite-capacity data queue located at the transmitter. A scheduler, which is oblivious to the current channel rates, selects one of the N channels for transmission. At the end of the slot, the scheduler only gets to know the transmission rate of the selected channel. The objective is to minimize the queue length regret, defined as the difference between the queue length at some time T achieved by an online policy and the queue length obtained by always transmitting over the single best channel in hindsight. We propose a weakly adaptive Multi-Armed Bandit (MAB) algorithm for minimizing the queue length regret in this setup. Unlike previous works, we do not make any stability assumptions about the queue or the arrival process. Hence, our result holds even when the queueing process is unstable. Our main observation is that the queue length regret can be upper bounded by the regret of a MAB policy that competes against the best channel in hindsight uniformly over all sub-intervals of [T]. As a technical contribution of independent interest, we then propose a weakly adaptive adversarial MAB policy which achieves Õ(√NT3/4) regret with high probability, implying the same bound for queue length regret. © 2025, CC BY-NC-ND.

关键词： Queueing theory

来源：评论

学校读者我要写书评

暂无评论

DARK-Adders: Digital Hardware Trojan Attack on Block-based Approximate Adders

DARK-Adders: Digital Hardware Trojan Attack on Block-based A...

引用

International Conference on VLSI Design

作者： Vishesh Mishra Neelofar Hassan Akshay Mehta Urbi Chatterjee Dept. of Computer Science and Engineering Indian Institute of Technology Kanpur Uttar Pradesh India Dept. of Electrical Engineering Indian Institute of Technology Kanpur Uttar Pradesh India

In recent times, approximate computing techniques have emerged as a popular computing paradigm that can significantly minimize consumption of resources at the cost of bounded loss in accuracy of results. More specifically, approximate arithmetic circuits such as addition and multiplication architecture have been widely deployed to benefit image processing, machine learning and other error-resilient applications. However, a major drawback attributed with approximate architectures is that the results generated by them are probabilistic in nature. This indeterminism in input-output characteristics results in security breaches during post-silicon validation. In other words, the relaxation in precision may lower the standard of testing metrics, thus making block-based approximate designs vulnerable to security attacks. This work proposes a digital Hardware Trojan Horse (HTH) that renders approximate adder circuits inefficient. The proposed attack exploits the probabilistic nature of approximate designs to successfully implant the HTH. The presented HTH replaces the original sum-bit generation logic in a 1-bit full adder widely present as a building block in a major class of block-based low latency approximate adders. Experimental results showcase that insertion of HTH in state-of-the-art approximate adders can reduce the end applications accuracy by 2–13%, and 24–31% when Trojan is inserted at Least significant Bit side (LSBs) and Most Significant Bit (MSBs) respectively. Our proposed Trojan does not add significant area overhead upon successful insertion. Additionally, the obtained results also signify that block-based low latency approximate adders with $N$ number of blocks are N/2 times more vulnerable to security attacks than their binary segmentation counterparts. Experimental results also demonstrate that the impact of a digital HTH may be considered to be relatively mild in exact computing designs, it proves to alter the intended behaviour of approximate circuit

关键词： computer architecture Machine learning Very large scale integration Probabilistic logic Hardware Trojan horses Security

来源：评论

学校读者我要写书评

暂无评论

Age of Information in a Two User MISO Interference Channel Under Rayleigh Fading

Age of Information in a Two User MISO Interference Channel U...

引用

National Conference on Communications (NCC)

作者： J Deeraj Kumar Parthajit Mohapatra Nikolaos Pappas Dept. of Electrical Engineering Indian Institute of Technology Tirupati Tirupati India Dept. of Computer and Information Science Linköping University Sweden

ISBN: (数字)9798331536015

ISBN: (纸本)9798331536022

This work explores the age of information (AoI) in a two-user Rayleigh faded multiple input single output (MISO) Gaussian interference channel. Each transmitter samples the data from a sensor and transmits the data with some random probability. The evolution of AoI is modeled using a Discrete Time Markov Chain (DTMC). Average AoI is characterized in the case of two prominent interference mitigation schemes namely treating interference as noise (TIN) and successive interference cancellation (SIC). Additionally, the impact of multiple transmit antennas and random access probability on the average AoI at the receiver is analyzed.

关键词： Interference cancellation Prevention and mitigation Transmitting antennas Receiving antennas Rayleigh channels Tin Information age Interference channels Decoding MISO

来源：评论

学校读者我要写书评

暂无评论

Pharmacy Location and Recommendation Services

Pharmacy Location and Recommendation Services

引用

IEEE International Conference on System, Computation, Automation and Networking (ICSCAN)

作者： R. Dhanalakshmi N. Vijayaraghavan R Valli Sahaya Beni Prathiba Centre for Cyber Physical Systems School of Computer Science and Engineering Vellore Institute of Technology Chennai Dept.of Mathematics KCG College of Technology Chennai India Dept.of Electronics and Communication Manakula Vinayagar Institute of Technology Puducherry India

The medical sector has continually advanced in recent years. The integration of medical services with a centralized system, on the other hand, is effective in enhancing people's lives thanks to the technology that is currently available. This system will help people to make out their medicinal requirements in their nearby location. If the exact medicine is not available alternative medicines are recommended based on user searched medicine combination it is achieved by using the rank-based algorithm to find similar medicine. This approach will computerize all pharmacy records so that customers and pharmacists can access them whenever and wherever they choose. The proposed system is to lay out a plan of action to the consumers plus pharmacist and do the representation of web services between them. This system intensifies the engagement of unorganized pharmacies into the computerized system to lift medical assistance. In such a way, the system will become the forthcoming way of e-medicinal services and advances the approach of medicinal assistance that are given to the patients by hospitals and doctors. It also helps to get drug availability in every locality

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Novel Gain Enhanced Folded Cascode OPAMP in 28nm CMOS technology

A Novel Gain Enhanced Folded Cascode OPAMP in 28nm CMOS Tech...

引用

2022 IEEE International Conference on Electrical, computer, and Energy Technologies, ICECET 2022

作者： Kumar, Ravi Nagulapalli, Rajasekhar Vishvakarma, S.K. Indian Institute of Technology Indore Dept. of Electrical Engineering Madhya Pradesh India Oxford Brookes University Dept. of Electrical and Computer Science Oxford United Kingdom

ISBN: (数字)9781665470872

ISBN: (纸本)9781665470872

Folded cascade Opamp is the preferred first stage choice in a typical two-stage opamp due to relaxed headroom and better output swing. Traditional Folded cascade opamp gain and noise are compromised due to the folding transistors. In this work, FVF assisted differential pair has been proposed which extracts the signal and injects it onto the gate of the folding devices, hence they will contribute gain and noise will be reduced. A prototype was developed in 2Snm TSMC CMOS technology and post-layout simulation results were performed. The proposed opamp achieves 68dB DC gain, which is 18dB higher than the conventional one. Also, the proposal exhibits 20% improvement in the input-referred noise for the same power consumption. Further, the circuit consumes 33W of power from a 1V power supply and occupies a 78∗46m2 silicon area. © 2022 IEEE.

关键词： Operational amplifiers

来源：评论

学校读者我要写书评

暂无评论

Text Transmission Through Li-Fi in WSN Using Smart Based Secure Morse Code Algorithm

Text Transmission Through Li-Fi in WSN Using Smart Based Sec...

引用

Multi-Agent Systems for Collaborative Intelligence (ICMSCI), International Conference on

作者： I. Vasudevan P. Malathi S. Siddik P.R Therasa M.Mohamed Nasurudeen R.M.Dilip Charaan Dept.of CSE Vel Tech Rangarajan Dr.Sagunthala R&D Institute of Science and Technology Chennai India Dept of Computer Science Saveetha College of Liberal Arts and Sciences SIMATS Dept.of EEE Pavaai Engineering College Namakkal India Dept.of CSE R.M.K Engineering College Kavaraipettai India Dept.of CSE V.S.B Engineering College Karur India

ISBN: (数字)9798331509828

ISBN: (纸本)9798331509835

Cyber security is one the most challenging task in today's world. Most of the existing Li-Fi technology in wireless sensor network is to detect the data within the short range of distance less accuracy. In this research, we proposed a Smart based Secure Morse Code (SSMC) technique. Here there are two part, first is encoding and next is decoding. This proposed encoding technique takes user input in text based form and convert these text to standard morse code pattern and decoding does in the opposite direction. The transmitter parts encode the user input and send the pattern through laser module. The receiver part receives the pattern sends to microcontroller and then decodes and displays the data. In our experiment results shows that the proposed technique can transmit data over long distance with help of morse code encoding decoding algorithm. This technique takes 5ms for the whole transmission and process the transmission accurately.

关键词： Wireless communication Wireless sensor networks Codes Transmitters Receivers Light fidelity Encoding Decoding computer crime Standards

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：