CAREER: A Dynamic Program Monitoring Framework Using Neural Network Hardware

职业：使用神经网络硬件的动态程序监控框架

基本信息

批准号：
1931078
负责人：
Abdullah Muzahid
金额：
$ 31.57万
依托单位：
Texas A&M Engineering Experiment Station
依托单位国家：
美国
项目类别：
Continuing Grant
财政年份：
2018
资助国家：
美国
起止时间：
2018-09-01 至 2023-03-31
项目状态：
已结题

来源：
https://www.nsf.gov/awardsearch/showAward?AWD_ID=1931078&HistoricalAwards=false
关键词：
CAREER Dynamic Program Monitoring Framework

项目摘要

Software bugs and security attacks cripple US economy by costing more than $150 billion a year. However, there has been no major innovation in this context. This research project aims to change that fact with the help of neural network based hardware. If the project is successful, it will significantly affect current industry practices and spur a new trend. It will encourage companies to invest in new techniques for debugging and security attack analysis using neural network hardware and make a compelling use case for the hardware implementation, thereby influencing continuous investment in neural network hardware. In addition, the project will contribute to the research and educational activities of a minority serving institution. Students will be tightly integrated into the project through dissertation, thesis work, and undergraduate research work. The PI will incorporate emerging architecture design and its programming in undergraduate and graduate coursework. Moreover, the PI will involve local high school students in computer science related projects through summer internships.Neural network is a machine learning technique that mimics human brain. Therefore, neural network hardware provides some unique capabilities that can be utilized in many different ways. This project proposes to utilize neural network hardware for "program monitoring". Program execution monitoring is often used to detect software bugs, performance issues, security attacks etc. Neural network hardware will learn the normal "behavior" of the program. Then it will detect any deviation of such behavior. Such deviation can be attributed to software bugs, performance issues or security attacks. The proposed approach provides a general framework for handling these issues. Due to online learning and testing capability of neural network hardware, the framework will be adaptive to any change in program inputs, code, and platforms.

软件错误和安全攻击每年耗资超过1500亿美元，削弱了美国经济。但是，在这种情况下，没有重大创新。该研究项目旨在借助基于神经网络的硬件来改变这一事实。如果该项目成功，它将极大地影响当前的行业实践并刺激新趋势。它将鼓励公司投资于使用神经网络硬件调试和安全攻击分析的新技术，并为硬件实施做出令人信服的用例，从而影响神经网络硬件的持续投资。此外，该项目将有助于少数派服务机构的研究和教育活动。通过论文，论文工作和本科研究工作将学生紧密整合到项目中。 PI将在本科和研究生课程中纳入新兴的建筑设计及其编程。此外，PI将通过暑期实习参与计算机科学与计算机科学相关的项目。神经网络是一种模仿人脑的机器学习技术。因此，神经网络硬件提供了一些独特的功能，可以通过许多不同的方式使用。该项目建议利用神经网络硬件进行“程序监视”。程序执行监视通常用于检测软件错误，性能问题，安全攻击等。神经网络硬件将学习程序的正常“行为”。然后，它将检测到这种行为的任何偏差。这种偏差可以归因于软件错误，性能问题或安全攻击。拟议的方法为处理这些问题提供了一个一般框架。由于神经网络硬件的在线学习和测试能力，该框架将适应程序输入，代码和平台的任何更改。

项目成果

期刊论文数量（7）

专著数量（0）

科研奖励数量（0）

会议论文数量（0）

专利数量（0）

XMeter: Finding Approximable Functions and Predicting Their Accuracy

DOI：
10.1109/tc.2020.3005083
发表时间：
2021-07
期刊：
IEEE Transactions on Computers
影响因子：
3.7
作者：
Riad Akram;Shantanu Mandal;A. Muzahid
通讯作者：
Riad Akram;Shantanu Mandal;A. Muzahid

WHISTLE: CPU Abstractions for Hardware and Software Memory Safety Invariants

WHISTLE：硬件和软件内存安全不变量的 CPU 抽象

DOI：
10.1109/tc.2022.3180990
发表时间：
2022
期刊：
IEEE Transactions on Computers
影响因子：
3.7
作者：
Kim, Sungkeun;Mahmud, Farabi;Huang, Jiayi;Majumder, Pritam;Tsai, Chia-Che;Muzahid, Abdullah;Kim, Eun Jung
通讯作者：
Kim, Eun Jung

SmartIndex: Learning to Index Caches to Improve Performance

SmartIndex：学习索引缓存以提高性能

DOI：
10.1109/lca.2023.3264478
发表时间：
2023
期刊：
IEEE Computer Architecture Letters
影响因子：
2.3
作者：
Weston, Kevin;Mahmud, Farabi;Janfaza, Vahid;Muzahid, Abdullah
通讯作者：
Muzahid, Abdullah

Post-Silicon Customization Using Deep Neural Networks

使用深度神经网络进行硅后定制

DOI：
发表时间：
2023
期刊：
International Conference on Architecture of Computing Ssstems
影响因子：
0
作者：
Weston, Kevin;Janfaza, Vahid;Taur, Abhishek;Mungra, Dhara;Kansal, Arnav;Zahran, Mohammed;Muzahid, Abdullah
通讯作者：
Muzahid, Abdullah

MERCURY: Accelerating DNN Training By Exploiting Input Similarity

DOI：
10.1109/hpca56546.2023.10071051
发表时间：
2021-10
期刊：
2023 IEEE International Symposium on High-Performance Computer Architecture (HPCA)
影响因子：
0
作者：
Vahid Janfaza;Kevin Weston;Moein Razavi;Shantanu Mandal;Farabi Mahmud;Alex Hilty;A. Muzahid
通讯作者：
Vahid Janfaza;Kevin Weston;Moein Razavi;Shantanu Mandal;Farabi Mahmud;Alex Hilty;A. Muzahid