Representation of adversarial examples in human and deep neural networks : an fMRI study인간과 심층신경망의 적대적 예제 표현 : 기능적 자기공명영상 연구

Cited 0 time in webofscience Cited 0 time in scopus
  • Hit : 180
  • Download : 0
The recent success of brain-inspired deep neural networks (DNNs) in solving complex, high-level visual tasks has led to rising expectations for their potential to match the human visual system. However, DNNs exhibit idiosyncrasies that suggest their visual representation and processing might be substantially different from human vision. One limitation of DNNs is that they are vulnerable to adversarial examples, input images on which subtle, carefully designed noises are added to fool a machine classifier. The robustness of the human visual system against adversarial examples is potentially of great importance as it could uncover a key mechanistic feature that machine vision is yet to incorporate. In this study, we compare the visual representations of white- and black-box adversarial examples in DNNs and humans by leveraging functional magnetic resonance imaging (fMRI). We find a significant difference in representation patterns for different types of adversarial examples for both humans and DNNs. However, human performance on categorical judgment is not degraded by noise regardless of the type unlike DNN. These results suggest that adversarial examples may be differentially represented in the human visual system, but unable to affect the perceptual experience.
Advisors
Kim, Dae-Shikresearcher김대식researcher
Description
한국과학기술원 :전기및전자공학부,
Publisher
한국과학기술원
Issue Date
2019
Identifier
325007
Language
eng
Description

학위논문(석사) - 한국과학기술원 : 전기및전자공학부, 2019.8,[iii, 28 p. :]

Keywords

deep neural network▼avisual processing▼aadversarial example▼afMRI▼acomputational model of the brain; 심층신경망▼a시각 처리▼a적대적 예제▼a기능적자기공명영상▼a뇌 계산 모델

URI
http://hdl.handle.net/10203/283055
Link
http://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=875349&flag=dissertation
Appears in Collection
EE-Theses_Master(석사논문)
Files in This Item
There are no files associated with this item.

qr_code

  • mendeley

    citeulike


rss_1.0 rss_2.0 atom_1.0