FMitF: Track I: Scalable and Quantitative Verification for Neural Network Analysis and Design
FMITF:第一轨:神经网络分析和设计的可扩展和定量验证
基本信息
- 批准号:2124039
- 负责人:
- 金额:$ 74.92万
- 依托单位:
- 依托单位国家:美国
- 项目类别:Standard Grant
- 财政年份:2021
- 资助国家:美国
- 起止时间:2021-10-01 至 2025-09-30
- 项目状态:未结题
- 来源:
- 关键词:
项目摘要
Neural Networks (NNs) have been successful in many areas including computer vision, speech recognition, and natural language processing. However, due to the increasing adoption of NNs in safety-critical and socially sensitive domains such as self-driving cars, robotics, computer security, criminal justice, and medical diagnosis, there is a pressing need for developing verification techniques that can provide guarantees about dependability and safety of NN applications. Formal-verification techniques can provide guarantees of correctness; however, existing approaches are not effective in analyzing real-world NNs with large numbers of neurons and complicated model structures. This project sets a comprehensive research agenda focusing on a holistic formal-verification framework for NNs that will provide a systematic and principled approach for developing dependable and safe NNs. It is intended to benefit major machine-learning applications such as autonomous driving and contribute to the leadership of the United States in software engineering and artificial intelligence. The research findings are being widely disseminated through open-source software packages, publications in premier conferences and journals, tutorials at teaching workshops, as well as specialized K-12 programs for exposing the young generation to the frontiers of software verification and machine-learning research. The team of researchers working on this project are integrating methods from the classical computing fields such as software engineering, automated verification, and formal methods to address the unique research challenges in the dependability and safety of NN applications. Specific research directions include 1) novel symbolic quantitative analysis techniques that provide sound results for establishing dependability and safety of the state-of-the-art NN models; 2) a set of effective system-level optimizations for computation/memory efficient NN verification with sufficient cross-framework portability and high verification efficiency; 3) advanced neural architecture design and training support for exploring and developing neural network models with verifiable robustness. The success of this research agenda is intended to enable a more complete and efficient software stack for improving the scalability of NN verification techniques and the robustness of NN applications.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.
神经网络(NNS)在许多领域都取得了成功,包括计算机视觉,语音识别和自然语言处理。但是,由于NNS在安全至关重要和社会敏感领域(例如自动驾驶汽车,机器人技术,计算机安全,刑事司法和医学诊断)的采用越来越多,因此有迫切需要开发验证技术,可以提供有关NN应用的可靠性和安全性的保证。正式的验证技术可以提供正确性的保证;但是,现有方法无效地分析具有大量神经元和复杂模型结构的现实世界中的NN。该项目设定了一项全面的研究议程,重点介绍了NNS的整体正式验证框架,该框架将为开发可靠和安全的NN提供系统和原则性的方法。它旨在使主要的机器学习应用程序受益,例如自动驾驶,并为美国在软件工程和人工智能方面的领导而做出贡献。通过开源软件包,高级会议和期刊的出版物,教学研讨会的教程以及专门的K-12计划,可以通过开源软件包,专门的K-12计划来广泛传播研究发现,以将年轻一代暴露于软件验证和机器学习研究的前沿。从事该项目的研究人员团队正在整合经典计算领域的方法,例如软件工程,自动验证和正式方法,以应对NN应用程序可靠性和安全性的独特研究挑战。特定的研究方向包括1)新型的符号定量分析技术,这些技术为确定最先进的NN模型的可靠性和安全性提供了合理的结果; 2)一组有效的系统级优化,用于计算/存储器有效验证,并具有足够的跨框架便携性和高验证效率; 3)高级神经体系结构设计和培训支持,用于探索和开发具有可验证鲁棒性的神经网络模型。该研究议程的成功旨在使更完整,更有效的软件堆栈,以改善NN验证技术的可扩展性和NN应用的鲁棒性。该奖项反映了NSF的法定任务,并被认为是值得通过基金会的知识分子和更广泛影响的评估审查审查标准来通过评估来通过评估来获得支持的。
项目成果
期刊论文数量(8)
专著数量(0)
科研奖励数量(0)
会议论文数量(0)
专利数量(0)
ZENO: A Type-based Optimization Framework for Zero Knowledge Neural Network Inference
- DOI:10.1145/3617232.3624852
- 发表时间:2024-04
- 期刊:
- 影响因子:0
- 作者:Boyuan Feng;Zheng Wang;Yuke Wang;Shu Yang;Yufei Ding
- 通讯作者:Boyuan Feng;Zheng Wang;Yuke Wang;Shu Yang;Yufei Ding
APNN-TC: Accelerating Arbitrary Precision Neural Networks on Ampere GPU Tensor Cores
- DOI:10.1145/3458817.3476157
- 发表时间:2021-06
- 期刊:
- 影响因子:0
- 作者:Boyuan Feng;Yuke Wang;Tong Geng;Ang Li;Yufei Ding
- 通讯作者:Boyuan Feng;Yuke Wang;Tong Geng;Ang Li;Yufei Ding
Faith: An Efficient Framework for Transformer Verification on GPUs
Faith:GPU 上 Transformer 验证的高效框架
- DOI:
- 发表时间:2022
- 期刊:
- 影响因子:0
- 作者:Feng, Boyuan;Tang, Tianqi;Wang, Yuke;Chen, Zhaodong;Wang, Zheng;Yang, Shu;Xie, Yuan;Ding, Yufei
- 通讯作者:Ding, Yufei
TC-GNN: Bridging Sparse GNN Computation and Dense Tensor Cores on GPUs
- DOI:
- 发表时间:2021-12
- 期刊:
- 影响因子:0
- 作者:Yuke Wang;Boyuan Feng;Zheng Wang;Guyue Huang;Yufei Ding
- 通讯作者:Yuke Wang;Boyuan Feng;Zheng Wang;Guyue Huang;Yufei Ding
QGTC: accelerating quantized graph neural networks via GPU tensor core
- DOI:10.1145/3503221.3508408
- 发表时间:2021-11
- 期刊:
- 影响因子:0
- 作者:Yuke Wang;Boyuan Feng;Yufei Ding
- 通讯作者:Yuke Wang;Boyuan Feng;Yufei Ding
{{
item.title }}
{{ item.translation_title }}
- DOI:
{{ item.doi }} - 发表时间:
{{ item.publish_year }} - 期刊:
- 影响因子:{{ item.factor }}
- 作者:
{{ item.authors }} - 通讯作者:
{{ item.author }}
数据更新时间:{{ journalArticles.updateTime }}
{{ item.title }}
- 作者:
{{ item.author }}
数据更新时间:{{ monograph.updateTime }}
{{ item.title }}
- 作者:
{{ item.author }}
数据更新时间:{{ sciAawards.updateTime }}
{{ item.title }}
- 作者:
{{ item.author }}
数据更新时间:{{ conferencePapers.updateTime }}
{{ item.title }}
- 作者:
{{ item.author }}
数据更新时间:{{ patent.updateTime }}
Tevfik Bultan其他文献
Tevfik Bultan的其他文献
{{
item.title }}
{{ item.translation_title }}
- DOI:
{{ item.doi }} - 发表时间:
{{ item.publish_year }} - 期刊:
- 影响因子:{{ item.factor }}
- 作者:
{{ item.authors }} - 通讯作者:
{{ item.author }}
{{ truncateString('Tevfik Bultan', 18)}}的其他基金
Collaborative Research: SHF: Small: Automated Quantitative Assessment of Testing Difficulty
合作研究:SHF:小型:测试难度自动定量评估
- 批准号:
2008660 - 财政年份:2020
- 资助金额:
$ 74.92万 - 项目类别:
Standard Grant
SHF: Medium: Collaborative Research: HUGS: Human-Guided Software Testing and Analysis for Scalable Bug Detection and Repair
SHF:中:协作研究:HUGS:用于可扩展错误检测和修复的人工引导软件测试和分析
- 批准号:
1901098 - 财政年份:2019
- 资助金额:
$ 74.92万 - 项目类别:
Continuing Grant
SHF: Small: Differential Policy Verification and Repair for Access Control in the Cloud
SHF:小型:云中访问控制的差异策略验证和修复
- 批准号:
1817242 - 财政年份:2018
- 资助金额:
$ 74.92万 - 项目类别:
Standard Grant
NSF Travel and Attendance Grant Proposal for ISSTA/SPIN 2017
NSF ISSTA/SPIN 2017 差旅和出勤补助金提案
- 批准号:
1741648 - 财政年份:2017
- 资助金额:
$ 74.92万 - 项目类别:
Standard Grant
EAGER: Collaborative Research: Leveraging Graph Databases for Incremental and Scalable Symbolic Analysis and Verification of Web Applications
EAGER:协作研究:利用图形数据库进行增量和可扩展的 Web 应用程序符号分析和验证
- 批准号:
1548848 - 财政年份:2015
- 资助金额:
$ 74.92万 - 项目类别:
Standard Grant
SHF: Small: Data Model Verification for Web Applications
SHF:小型:Web 应用程序的数据模型验证
- 批准号:
1423623 - 财政年份:2014
- 资助金额:
$ 74.92万 - 项目类别:
Standard Grant
TC: Small: Collaborative Research: Viewpoints: Discovering Client- and Server-side Input Validation Inconsistencies to Improve Web Application Security
TC:小型:协作研究:观点:发现客户端和服务器端输入验证不一致以提高 Web 应用程序安全性
- 批准号:
1116967 - 财政年份:2011
- 资助金额:
$ 74.92万 - 项目类别:
Standard Grant
SHF: Small: Collaborative Research: Formal Analysis of Distributed Interactions
SHF:小型:协作研究:分布式交互的形式分析
- 批准号:
1117708 - 财政年份:2011
- 资助金额:
$ 74.92万 - 项目类别:
Standard Grant
TC: Small:Automata Based String Analysis for Detecting Vulnerabilities in Web Applications
TC:Small:基于自动机的字符串分析,用于检测 Web 应用程序中的漏洞
- 批准号:
0916112 - 财政年份:2009
- 资助金额:
$ 74.92万 - 项目类别:
Standard Grant
SoD-HCER: Design for Verification
SoD-HCER:验证设计
- 批准号:
0614002 - 财政年份:2006
- 资助金额:
$ 74.92万 - 项目类别:
Standard Grant
相似国自然基金
石羊河上游径流水源追踪量化的模拟研究
- 批准号:42301153
- 批准年份:2023
- 资助金额:30 万元
- 项目类别:青年科学基金项目
面向复杂场景的说话人追踪关键技术研究
- 批准号:62306029
- 批准年份:2023
- 资助金额:30 万元
- 项目类别:青年科学基金项目
用户兴趣迁移现象下基于图神经网络的舆情追踪技术研究
- 批准号:62302199
- 批准年份:2023
- 资助金额:30 万元
- 项目类别:青年科学基金项目
单波段机载LiDAR测深的瞬时海面确定及光线追踪
- 批准号:42304051
- 批准年份:2023
- 资助金额:30 万元
- 项目类别:青年科学基金项目
基于量子电压动态追踪补偿的精密磁通测量方法研究
- 批准号:52307021
- 批准年份:2023
- 资助金额:30 万元
- 项目类别:青年科学基金项目
相似海外基金
RII Track-4: NSF: Scalable MPI with Adaptive Compression for GPU-based Computing Systems
RII Track-4:NSF:适用于基于 GPU 的计算系统的具有自适应压缩的可扩展 MPI
- 批准号:
2327266 - 财政年份:2024
- 资助金额:
$ 74.92万 - 项目类别:
Standard Grant
RII Track-4: NSF: Extracting Pan Genomic Information from Metagenomic Data: Distributed Algorithms and Scalable Software
RII Track-4:NSF:从宏基因组数据中提取泛基因组信息:分布式算法和可扩展软件
- 批准号:
2327456 - 财政年份:2024
- 资助金额:
$ 74.92万 - 项目类别:
Standard Grant
ExpandQISE: Track 1: Harnessing a scalable platform to demonstrate multipartite quantum effects under strict conditions
ExpandQISE:轨道 1:利用可扩展平台在严格条件下演示多部分量子效应
- 批准号:
2328800 - 财政年份:2023
- 资助金额:
$ 74.92万 - 项目类别:
Standard Grant
FMiTF: Track-2 : Rigorous and Scalable Formal Floating-Point Error Analysis from LLVM
FMiTF:Track-2:来自 LLVM 的严格且可扩展的形式浮点误差分析
- 批准号:
2319507 - 财政年份:2023
- 资助金额:
$ 74.92万 - 项目类别:
Standard Grant
Harnessing Network Science to Personalize Scalable Interventions for Adolescent Depression
利用网络科学对青少年抑郁症进行个性化的可扩展干预措施
- 批准号:
10860020 - 财政年份:2023
- 资助金额:
$ 74.92万 - 项目类别: