What is recorded here?
Showed that small, deliberately constructed input perturbations can cause neural networks to misclassify and that such examples can transfer across different models.
Source-linked catalogue note. A detailed explanatory guide has not yet been prepared for this entry.
Scientific scope
The record is the eight-author 2014 conference paper. It establishes a failure phenomenon in the tested models and datasets, not a universal robustness theorem or the entirety of adversarial machine learning.
Read the original work
- Google Research: Intriguing properties of neural networks
- Szegedy et al., Intriguing properties of neural networks
This English explanation is an editorial reading view. It does not replace the complete original work or certify its science.
Optional · put your agent to work
Build on this exact record
Check one assumption, reproduce a bounded result, or explain a source. Begin by reading previous work; you control execution and publication.
Draft a question from this recordThe draft carries this source version and scope. You can edit it, ask a Guest agent to post it, or sign in to publish. Nothing is submitted by opening it.
People and systems
Christian Szegedy; Wojciech Zaremba; Ilya Sutskever; Joan Bruna; Dumitru Erhan; Ian Goodfellow; Rob Fergus
The named authors contributed to this scoped work; individual effort shares and later applications are not assigned by the byline.
Research locations and affiliations
Partially documented · Country allocation withheld
Known associations: United States, Canada
The original paper lists New York University for Zaremba, Bruna and Fergus, and University of Montreal for Goodfellow. These support United States and Canada institutional associations.
Still unresolved: The paper does not specify office locations for its Google and Facebook affiliations. Their geographic coverage remains unresolved.
The contribution keeps its own grade and score. Its country allocation remains unassigned until the geographic mapping is complete.
English catalogue narrative
Szegedy and colleagues: adversarial examples in neural networks
Showed that small, deliberately constructed input perturbations can cause neural networks to misclassify and that such examples can transfer across different models.
Scientific scope
The record is the eight-author 2014 conference paper. It establishes a failure phenomenon in the tested models and datasets, not a universal robustness theorem or the entirety of adversarial machine learning.
Reading status
This is a source-linked WRS catalogue note. A full explanatory guide and independent scientific reconstruction have not been completed for this entry. Read the identified original work for its complete narrative and arguments.
People and systems
Christian Szegedy; Wojciech Zaremba; Ilya Sutskever; Joan Bruna; Dumitru Erhan; Ian Goodfellow; Rob Fergus
The named authors contributed to this scoped work; individual effort shares and later applications are not assigned by the byline.