What the AI Saw

My student Anne Linja presented her poster “What the AI Saw: Examining human predictions of deep image classification errors” at the Midwest Cognitive Science Conference in Columbus OH May 28, 2019. A link to the poster is here:

Anne Linja, Lamia Alam, Shane T. Mueller: What the AI Saw: Examining human predictions of deep image classification errors. Midwest Cognitive Science, Columbus, OH, 2019.

Abstract

Deep Image classifiers have made amazing advances in both basic and applied problems in recent years. Nevertheless, they are still very limited and can be foiled by even simple image distortions. Importantly, the way they fail is often unexpected, and sometimes difficult to even understand. Thus, advances in image classifiers made to improve their transparency and the predictability of their errors may make more of a difference than algorithmic improvements that reduce error rates on benchmarks. To understand the types of expectations humans may have, we conducted a study in which students were asked to predict whether a generic AI system would correctly identify 10 classes of tools (axe, hammer, wrench, flashlight, pliers, saw, scissors, screwdriver, tape measure), with a variety of image transforms (e.g., borders, outline filters, additional objects inserted into image, etc.), and also examined how five commercial deep image classifiers performed on the same imagery. Results revealed that humans tended to predict that distortions and distractions would lead to impairment of the AI systems, and although AI failures did incorporate these factors, they also involved many class-level errors (e.g., calling a wrench a tool or a product), and feature-errors (calling a hammer 'metal' or 'wood') not identified by human novice users. Results will be discussed in the context of Explainable AI systems.

BibTeX (Download)

@conference{Linja2019,
title = {What the AI Saw: Examining human predictions of deep image classification errors},
author = {Anne Linja and Lamia Alam and Shane T. Mueller},
url = {https://www.researchgate.net/publication/333246714_What_the_AI_saw_Examining_human_predictions_of_deep_image_classification_errors
},
year  = {2019},
date = {2019-05-25},
booktitle = {Midwest Cognitive Science},
address = {Columbus, OH},
abstract = {Deep Image classifiers have made amazing advances in both basic and applied problems in recent years. Nevertheless, they are still very limited and can be foiled by even simple image distortions. Importantly, the way they fail is often unexpected, and sometimes difficult to even understand. Thus, advances in image classifiers made to improve their transparency and the predictability of their errors may make more of a difference than algorithmic improvements that reduce error rates on benchmarks. To understand the types of expectations humans may have, we conducted a study in which students were asked to predict whether a generic AI system would correctly identify 10 classes of tools (axe, hammer, wrench, flashlight, pliers, saw, scissors, screwdriver, tape measure), with a variety of image transforms (e.g., borders, outline filters, additional objects inserted into image, etc.), and also examined how five commercial deep image classifiers performed on the same imagery. Results revealed that humans tended to predict that distortions and distractions would lead to impairment of the AI systems, and although AI failures did incorporate these factors, they also involved many class-level errors (e.g., calling a wrench a tool or a product), and feature-errors (calling a hammer 'metal' or 'wood') not identified by human novice users. Results will be discussed in the context of Explainable AI systems.},
keywords = {XAI},
pubstate = {published},
tppubtype = {conference}
}

98 comments on “What the AI Saw

  1. Pingback: order cialis
  2. Pingback: tadalafil
  3. Pingback: viagra for sale
  4. Pingback: cialis 10mg
  5. Pingback: viagra mg 25
  6. Pingback: meritking
  7. Pingback: elexusbet
  8. Pingback: eurocasino
  9. Pingback: madridbet
  10. Pingback: meritroyalbet
  11. Pingback: eurocasino
  12. Pingback: eurocasino
  13. Pingback: meritroyalbet
  14. Pingback: pokies lucky 88
  15. Pingback: viagra pill 50mg
  16. Pingback: cheap viagra soft
  17. Pingback: cheapest viagra
  18. Pingback: mazhor4sezon
  19. Pingback: filmfilmfilmes
  20. Pingback: gRh9UPV
  21. Pingback: 9-05-2022
  22. Pingback: kinoteatrzarya.ru
  23. Pingback: TopGun2022
  24. Pingback: Xvideos
  25. Pingback: XVIDEOSCOM Videos
  26. Pingback: ivanesva
  27. Pingback: Netflix
  28. Pingback: FILM
  29. Pingback: designchita.ru
  30. Pingback: YA-krasneyu
  31. Pingback: design-human.ru
  32. Pingback: designmsu.ru
  33. Pingback: vkl-design.ru
  34. Pingback: irida-design.ru
  35. Pingback: projectio
  36. Pingback: psy online
  37. Pingback: Gz92uNNH
  38. Pingback: do-posle-psihologa
  39. Pingback: uels ukrain
  40. Pingback: DPTPtNqS
  41. Pingback: qQ8KZZE6
  42. Pingback: D6tuzANh
  43. Pingback: SHKALA TONOV
  44. Pingback: chelovek-iz-90-h
  45. Pingback: 3Hk12Bl
  46. Pingback: 3NOZC44
  47. Pingback: 01211
  48. Pingback: tor-lyubov-i-grom
  49. Pingback: film-tor-2022
  50. Pingback: hd-tor-2022
  51. Pingback: hdorg2.ru
  52. Pingback: JXNhGmmt
  53. Pingback: Psikholog
  54. Pingback: netstate.ru
  55. Pingback: Link
  56. Pingback: psy
  57. Pingback: bit.ly
  58. Pingback: cleantalkorg2.ru
  59. Pingback: bucha killings
  60. Pingback: War in Ukraine
  61. Pingback: Ukraine
  62. Pingback: site
  63. Pingback: stats
  64. Pingback: Ukraine-war
  65. Pingback: movies
  66. Pingback: gidonline
  67. Pingback: web
  68. Pingback: film.8filmov.ru
  69. Pingback: video
  70. Pingback: filmgoda.ru
  71. Pingback: rodnoe-kino-ru
  72. Pingback: confeitofilm
  73. Pingback: stat.netstate.ru
  74. Pingback: sY5am

Comments are closed.