为何深度学习人工智能如此容易被欺骗。 - Suppr

Suppr

超能文献

Why deep-learning AIs are so easy to fool.

作者信息

Heaven Douglas

出版信息

Nature. 2019 Oct;574(7777):163-166. doi: 10.1038/d41586-019-03013-5.

DOI:10.1038/d41586-019-03013-5

PMID:31597977

Abstract

摘要

相似文献

Why deep-learning AIs are so easy to fool.

Nature. 2019 Oct;574(7777):163-166. doi: 10.1038/d41586-019-03013-5.

Image denoising using deep CNN with batch renormalization.

Neural Netw. 2020 Jan;121:461-473. doi: 10.1016/j.neunet.2019.08.022. Epub 2019 Sep 5.

Recognition and Repetition Counting for ComplexPhysical Exercises with Deep Learning.

Sensors (Basel). 2019 Feb 10;19(3):714. doi: 10.3390/s19030714.

3D-Convolutional Neural Network with Generative Adversarial Network and Autoencoder for Robust Anomaly Detection in Video Surveillance.

Int J Neural Syst. 2020 Jun;30(6):2050034. doi: 10.1142/S0129065720500343. Epub 2020 May 28.

BoSR: A CNN-based aurora image retrieval method.

Neural Netw. 2019 Aug;116:188-197. doi: 10.1016/j.neunet.2019.04.012. Epub 2019 May 2.

Parametric Deformable Exponential Linear Units for deep neural networks.

Neural Netw. 2020 May;125:281-289. doi: 10.1016/j.neunet.2020.02.012. Epub 2020 Feb 26.

Multiple Discrimination and Pairwise CNN for view-based 3D object retrieval.

Neural Netw. 2020 May;125:290-302. doi: 10.1016/j.neunet.2020.02.017. Epub 2020 Feb 29.

Deep Manifold Learning Combined With Convolutional Neural Networks for Action Recognition.

IEEE Trans Neural Netw Learn Syst. 2018 Sep;29(9):3938-3952. doi: 10.1109/TNNLS.2017.2740318. Epub 2017 Sep 15.

Deeply Learned View-Invariant Features for Cross-View Action Recognition.

IEEE Trans Image Process. 2017 Jun;26(6):3028-3037. doi: 10.1109/TIP.2017.2696786. Epub 2017 Apr 24.

An adaptive deep Q-learning strategy for handwritten digit recognition.

Neural Netw. 2018 Nov;107:61-71. doi: 10.1016/j.neunet.2018.02.010. Epub 2018 Feb 22.

引用本文的文献

Gender- and Age-Associated Variations in the Prevalence of Atelectasis, Effusion, and Nodules on Chest Radiographs: A Large-Scale Analysis Using the NIH ChestX-Ray8.

Diagnostics (Basel). 2025 May 26;15(11):1330. doi: 10.3390/diagnostics15111330.

Comprehensive reporting guidelines and checklist for studies developing and utilizing artificial intelligence models.

Korean J Anesthesiol. 2025 Jun;78(3):199-214. doi: 10.4097/kja.25075. Epub 2025 Mar 26.

Human control of AI systems: from supervision to teaming.

AI Ethics. 2025;5(2):1535-1548. doi: 10.1007/s43681-024-00489-4. Epub 2024 May 28.

Invisible CMOS Camera Dazzling for Conducting Adversarial Attacks on Deep Neural Networks.

Sensors (Basel). 2025 Apr 4;25(7):2301. doi: 10.3390/s25072301.

Closing the Automation Gap: Robust AI for Dual-Stain Cervical Cancer Screening Triage.

Res Sq. 2025 Mar 4:rs.3.rs-5985837. doi: 10.21203/rs.3.rs-5985837/v1.

Comparison of the impact of rectal susceptibility artifacts in prostate magnetic resonance imaging on subjective evaluation and deep learning: a two-center retrospective study.

BMC Med Imaging. 2025 Feb 25;25(1):61. doi: 10.1186/s12880-025-01602-7.

High-reward, high-risk technologies? An ethical and legal account of AI development in healthcare.

BMC Med Ethics. 2025 Jan 15;26(1):4. doi: 10.1186/s12910-024-01158-1.

Long-Term Exposure to Outdoor Ultrafine Particles and Black Carbon and Effects on Mortality in Montreal and Toronto, Canada.

Res Rep Health Eff Inst. 2024 Jul;2024(217):1-63.

An Integrated Machine Learning Approach for Congestive Heart Failure Prediction.

Diagnostics (Basel). 2024 Mar 29;14(7):736. doi: 10.3390/diagnostics14070736.

Exploring dermoscopic structures for melanoma lesions' classification.

Front Big Data. 2024 Mar 25;7:1366312. doi: 10.3389/fdata.2024.1366312. eCollection 2024.

本文引用的文献

Adversarial attacks on medical machine learning.

Science. 2019 Mar 22;363(6433):1287-1289. doi: 10.1126/science.aaw4399.

A general reinforcement learning algorithm that masters chess, shogi, and Go through self-play.

Science. 2018 Dec 7;362(6419):1140-1144. doi: 10.1126/science.aar6404.

文献AI研究员

20分钟写一篇综述，助力文献阅读效率提升50倍。

立即体验

用中文搜PubMed

大模型驱动的PubMed中文搜索引擎

马上搜索

文档翻译

学术文献翻译模型，支持多种主流文档格式。

立即体验

Why deep-learning AIs are so easy to fool.

作者信息

Heaven Douglas

出版信息

Nature. 2019 Oct;574(7777):163-166. doi: 10.1038/d41586-019-03013-5.

DOI:10.1038/d41586-019-03013-5

PMID:31597977

Abstract

摘要

相似文献

Why deep-learning AIs are so easy to fool.

Nature. 2019 Oct;574(7777):163-166. doi: 10.1038/d41586-019-03013-5.

Image denoising using deep CNN with batch renormalization.

Neural Netw. 2020 Jan;121:461-473. doi: 10.1016/j.neunet.2019.08.022. Epub 2019 Sep 5.

Recognition and Repetition Counting for ComplexPhysical Exercises with Deep Learning.

Sensors (Basel). 2019 Feb 10;19(3):714. doi: 10.3390/s19030714.

3D-Convolutional Neural Network with Generative Adversarial Network and Autoencoder for Robust Anomaly Detection in Video Surveillance.

Int J Neural Syst. 2020 Jun;30(6):2050034. doi: 10.1142/S0129065720500343. Epub 2020 May 28.

BoSR: A CNN-based aurora image retrieval method.

Neural Netw. 2019 Aug;116:188-197. doi: 10.1016/j.neunet.2019.04.012. Epub 2019 May 2.

Parametric Deformable Exponential Linear Units for deep neural networks.

Neural Netw. 2020 May;125:281-289. doi: 10.1016/j.neunet.2020.02.012. Epub 2020 Feb 26.

Multiple Discrimination and Pairwise CNN for view-based 3D object retrieval.

Neural Netw. 2020 May;125:290-302. doi: 10.1016/j.neunet.2020.02.017. Epub 2020 Feb 29.

Deep Manifold Learning Combined With Convolutional Neural Networks for Action Recognition.

IEEE Trans Neural Netw Learn Syst. 2018 Sep;29(9):3938-3952. doi: 10.1109/TNNLS.2017.2740318. Epub 2017 Sep 15.

Deeply Learned View-Invariant Features for Cross-View Action Recognition.

IEEE Trans Image Process. 2017 Jun;26(6):3028-3037. doi: 10.1109/TIP.2017.2696786. Epub 2017 Apr 24.

An adaptive deep Q-learning strategy for handwritten digit recognition.

Neural Netw. 2018 Nov;107:61-71. doi: 10.1016/j.neunet.2018.02.010. Epub 2018 Feb 22.

引用本文的文献

Gender- and Age-Associated Variations in the Prevalence of Atelectasis, Effusion, and Nodules on Chest Radiographs: A Large-Scale Analysis Using the NIH ChestX-Ray8.

Diagnostics (Basel). 2025 May 26;15(11):1330. doi: 10.3390/diagnostics15111330.

Comprehensive reporting guidelines and checklist for studies developing and utilizing artificial intelligence models.

Korean J Anesthesiol. 2025 Jun;78(3):199-214. doi: 10.4097/kja.25075. Epub 2025 Mar 26.

Human control of AI systems: from supervision to teaming.

AI Ethics. 2025;5(2):1535-1548. doi: 10.1007/s43681-024-00489-4. Epub 2024 May 28.

Invisible CMOS Camera Dazzling for Conducting Adversarial Attacks on Deep Neural Networks.

Sensors (Basel). 2025 Apr 4;25(7):2301. doi: 10.3390/s25072301.

Closing the Automation Gap: Robust AI for Dual-Stain Cervical Cancer Screening Triage.

Res Sq. 2025 Mar 4:rs.3.rs-5985837. doi: 10.21203/rs.3.rs-5985837/v1.

Comparison of the impact of rectal susceptibility artifacts in prostate magnetic resonance imaging on subjective evaluation and deep learning: a two-center retrospective study.

BMC Med Imaging. 2025 Feb 25;25(1):61. doi: 10.1186/s12880-025-01602-7.

High-reward, high-risk technologies? An ethical and legal account of AI development in healthcare.

BMC Med Ethics. 2025 Jan 15;26(1):4. doi: 10.1186/s12910-024-01158-1.

Long-Term Exposure to Outdoor Ultrafine Particles and Black Carbon and Effects on Mortality in Montreal and Toronto, Canada.

Res Rep Health Eff Inst. 2024 Jul;2024(217):1-63.

An Integrated Machine Learning Approach for Congestive Heart Failure Prediction.

Diagnostics (Basel). 2024 Mar 29;14(7):736. doi: 10.3390/diagnostics14070736.

Exploring dermoscopic structures for melanoma lesions' classification.

Front Big Data. 2024 Mar 25;7:1366312. doi: 10.3389/fdata.2024.1366312. eCollection 2024.

本文引用的文献

Adversarial attacks on medical machine learning.

Science. 2019 Mar 22;363(6433):1287-1289. doi: 10.1126/science.aaw4399.

A general reinforcement learning algorithm that masters chess, shogi, and Go through self-play.

Science. 2018 Dec 7;362(6419):1140-1144. doi: 10.1126/science.aar6404.