Defense against adversarial attacks on deep learning in computer vision
防御计算机视觉深度学习的对抗性攻击
基本信息
- 批准号:DP190102443
- 负责人:
- 金额:$ 29.92万
- 依托单位:
- 依托单位国家:澳大利亚
- 项目类别:Discovery Projects
- 财政年份:2019
- 资助国家:澳大利亚
- 起止时间:2019-01-01 至 2023-12-31
- 项目状态:已结题
- 来源:
- 关键词:
项目摘要
Computer vision applications rely heavily on deep learning, which is highly vulnerable to being fooled by adding subtle perturbations to object/image textures that are imperceptible to humans. This project aims to develop defense mechanisms to detect and remove adversarial patterns from the input images. The project expects to advance knowledge in understanding the vulnerabilities of deep learning, and to design deep learning architectures that are inherently robust. The outcomes of this project will increase the security and reliability of computer vision by detecting, reporting and nullifying such attacks and will benefit the general public and industry on many fronts.
计算机视觉应用在很大程度上依赖于深度学习,而深度学习很容易被人类无法察觉的物体/图像纹理添加微妙的扰动所欺骗。该项目旨在开发防御机制,以从输入图像中检测和去除对抗模式。该项目希望在理解深度学习漏洞方面取得进展,并设计出固有鲁棒性的深度学习架构。该项目的成果将通过检测、报告和消除此类攻击来提高计算机视觉的安全性和可靠性,并将在许多方面使公众和工业界受益。
项目成果
期刊论文数量(0)
专著数量(0)
科研奖励数量(0)
会议论文数量(0)
专利数量(0)
数据更新时间:{{ journalArticles.updateTime }}
{{
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 }}
Prof Ajmal Mian其他文献
Prof Ajmal Mian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('Prof Ajmal Mian', 18)}}的其他基金
3D Diffusion Models for Generating and Understanding 3D Scenes
用于生成和理解 3D 场景的 3D 扩散模型
- 批准号:
DP240101926 - 财政年份:2024
- 资助金额:
$ 29.92万 - 项目类别:
Discovery Projects
Defending Artificial Intelligence against deception attacks
保护人工智能免受欺骗攻击
- 批准号:
NS220100007 - 财政年份:2022
- 资助金额:
$ 29.92万 - 项目类别:
National Intelligence and Security Discovery Research Grants
Robust and Explainable 3D Computer Vision
稳健且可解释的 3D 计算机视觉
- 批准号:
FT210100268 - 财政年份:2022
- 资助金额:
$ 29.92万 - 项目类别:
ARC Future Fellowships
View and shape invariant modeling of human actions for smart surveillance
查看和塑造用于智能监控的人类行为的不变模型
- 批准号:
DP160101458 - 财政年份:2016
- 资助金额:
$ 29.92万 - 项目类别:
Discovery Projects
Machine Learning for Fracture Risk Assessment from Simple Radiography
机器学习通过简单射线照相进行骨折风险评估
- 批准号:
LP150101052 - 财政年份:2016
- 资助金额:
$ 29.92万 - 项目类别:
Linkage Projects
Active multispectral computer vision for defence and security
用于国防和安全的主动多光谱计算机视觉
- 批准号:
DP110102399 - 财政年份:2011
- 资助金额:
$ 29.92万 - 项目类别:
Discovery Projects
Integration of Spatiotemporal Video Data for Realtime Smart Proactive Surveillance
时空视频数据整合,实现实时智能主动监控
- 批准号:
DP0881813 - 财政年份:2008
- 资助金额:
$ 29.92万 - 项目类别:
Discovery Projects
相似海外基金
Robust Defences against Adversarial Machine Learning for UAV Systems
针对无人机系统对抗性机器学习的稳健防御
- 批准号:
LP230100083 - 财政年份:2024
- 资助金额:
$ 29.92万 - 项目类别:
Linkage Projects
Precision Design of Antimicrobial Peptides Against Bacterial Infections
抗细菌感染抗菌肽的精密设计
- 批准号:
10708842 - 财政年份:2022
- 资助金额:
$ 29.92万 - 项目类别:
Boosting Robustness of Deep Neural Networks against Sparsity-aware Adversarial Attacks
提高深度神经网络对抗稀疏感知对抗攻击的鲁棒性
- 批准号:
580570-2022 - 财政年份:2022
- 资助金额:
$ 29.92万 - 项目类别:
Alliance Grants
EXCELLENCE in RESEARCH: SECURING MACHINE LEARNING AGAINST ADVERSARIAL ATTACKS FOR CONNECTED AND AUTONOMOUS VEHICLES
卓越的研究:保护联网和自动驾驶车辆的机器学习免受对抗性攻击
- 批准号:
2200457 - 财政年份:2022
- 资助金额:
$ 29.92万 - 项目类别:
Standard Grant
Precision Design of Antimicrobial Peptides Against Bacterial Infections
抗细菌感染抗菌肽的精密设计
- 批准号:
10522451 - 财政年份:2022
- 资助金额:
$ 29.92万 - 项目类别:
Rapid response for pandemics: single cell sequencing and deep learning to predict antibody sequences against an emerging antigen
快速应对流行病:单细胞测序和深度学习预测针对新兴抗原的抗体序列
- 批准号:
10845715 - 财政年份:2021
- 资助金额:
$ 29.92万 - 项目类别:
Learning Internal Representations Robust against Adversarial Attacks
学习抵御对抗性攻击的内部表示
- 批准号:
20K19824 - 财政年份:2020
- 资助金额:
$ 29.92万 - 项目类别:
Grant-in-Aid for Early-Career Scientists
SaTC: CORE: Small: Attack-Agnostic Defenses against Adversarial Inputs in Learning Systems
SaTC:核心:小:针对学习系统中的对抗性输入的与攻击无关的防御
- 批准号:
1953813 - 财政年份:2019
- 资助金额:
$ 29.92万 - 项目类别:
Standard Grant
SaTC: CORE: Medium: Securing the Voice Processing Pipeline Against Adversarial Audio
SaTC:核心:中:保护语音处理管道免受对抗性音频的影响
- 批准号:
1933208 - 财政年份:2019
- 资助金额:
$ 29.92万 - 项目类别:
Standard Grant
EAGER: Invisible Shield: Can Compression Harden Deep Neural Networks Universally Against Adversarial Attacks?
EAGER:隐形盾牌:压缩能否使深层神经网络普遍抵御对抗性攻击?
- 批准号:
2011260 - 财政年份:2019
- 资助金额:
$ 29.92万 - 项目类别:
Standard Grant