full transcript

From the Ted Talk by Sebastian Thrun and Chris Anderson: What AI is -- and isn't


Unscramble the Blue Letters


ST: This was last Thursday. That's a moving piece. What we've shwon before and we published in "Nature" earlier this year was this idea that we show doetrsagtomils images and our computer program images, and count how often they're right. But all these images are past images. They've all been biopsied to make sure we had the correct cacsifsoliaitn. This one wasn't. This one was actually done at Stanford by one of our collaborators. The sorty goes that our crotllbaoaor, who is a world-famous dgorsmetloiat, one of the three best, apparently, looked at this mole and said, "This is not skin cancer." And then he had a second mneomt, where he said, "Well, let me just check with the app." So he took out his iPhone and ran our piece of software, our "pocket dermatologist," so to speak, and the iPhone said: cancer. It said melanoma. And then he was confused. And he decided, "OK, maybe I trust the iPhone a little bit more than myself," and he sent it out to the lab to get it biopsied. And it came up as an aggressive melanoma. So I think this might be the first time that we actually found, in the practice of using deep leanrnig, an atuacl person whose melanoma would have gone unclassified, had it not been for deep learning.

Open Cloze


ST: This was last Thursday. That's a moving piece. What we've _____ before and we published in "Nature" earlier this year was this idea that we show ______________ images and our computer program images, and count how often they're right. But all these images are past images. They've all been biopsied to make sure we had the correct ______________. This one wasn't. This one was actually done at Stanford by one of our collaborators. The _____ goes that our ____________, who is a world-famous _____________, one of the three best, apparently, looked at this mole and said, "This is not skin cancer." And then he had a second ______, where he said, "Well, let me just check with the app." So he took out his iPhone and ran our piece of software, our "pocket dermatologist," so to speak, and the iPhone said: cancer. It said melanoma. And then he was confused. And he decided, "OK, maybe I trust the iPhone a little bit more than myself," and he sent it out to the lab to get it biopsied. And it came up as an aggressive melanoma. So I think this might be the first time that we actually found, in the practice of using deep ________, an ______ person whose melanoma would have gone unclassified, had it not been for deep learning.

Solution


  1. dermatologist
  2. classification
  3. shown
  4. collaborator
  5. moment
  6. learning
  7. dermatologists
  8. story
  9. actual

Original Text


ST: This was last Thursday. That's a moving piece. What we've shown before and we published in "Nature" earlier this year was this idea that we show dermatologists images and our computer program images, and count how often they're right. But all these images are past images. They've all been biopsied to make sure we had the correct classification. This one wasn't. This one was actually done at Stanford by one of our collaborators. The story goes that our collaborator, who is a world-famous dermatologist, one of the three best, apparently, looked at this mole and said, "This is not skin cancer." And then he had a second moment, where he said, "Well, let me just check with the app." So he took out his iPhone and ran our piece of software, our "pocket dermatologist," so to speak, and the iPhone said: cancer. It said melanoma. And then he was confused. And he decided, "OK, maybe I trust the iPhone a little bit more than myself," and he sent it out to the lab to get it biopsied. And it came up as an aggressive melanoma. So I think this might be the first time that we actually found, in the practice of using deep learning, an actual person whose melanoma would have gone unclassified, had it not been for deep learning.

Frequently Occurring Word Combinations


ngrams of length 2

collocation frequency
machine learning 10
deep learning 5
software engineer 3
million games 3
flying cars 3
artificial intelligence 2
software engineers 2
lines long 2
million lines 2
billion web 2
web pages 2
san francisco 2
traffic lights 2
human dermatologists 2
massive amounts 2
medical supply 2
side effects 2
issue anymore 2

ngrams of length 3

collocation frequency
billion web pages 2


Important Words


  1. actual
  2. aggressive
  3. app
  4. apparently
  5. biopsied
  6. bit
  7. cancer
  8. check
  9. classification
  10. collaborator
  11. collaborators
  12. computer
  13. confused
  14. correct
  15. count
  16. decided
  17. deep
  18. dermatologist
  19. dermatologists
  20. earlier
  21. idea
  22. images
  23. iphone
  24. lab
  25. learning
  26. looked
  27. melanoma
  28. mole
  29. moment
  30. moving
  31. person
  32. piece
  33. practice
  34. program
  35. published
  36. ran
  37. show
  38. shown
  39. skin
  40. software
  41. speak
  42. stanford
  43. story
  44. thursday
  45. time
  46. trust
  47. unclassified
  48. year