back to article Imaginary numbers help AIs solve the very real problem of adversarial imagery

Boffins from Duke University say they have figured out a way to help protect artificial intelligences from adversarial image-modification attacks: by throwing a few imaginary numbers their way. Computer vision systems which recognise objects are at the heart of a whole swathe of shiny new technologies, from automated shops to …

  1. Anonymous Coward
    Anonymous Coward

    Can't resist...

    Imaginary numbers? So they really are making it up as they go along...

    (Yes, I know. It's more complex than that...)

    1. steelpillow Silver badge

      Re: Can't resist...

      I am a little phased by the magnitude of your insinuations.

      1. Paul Crawford Silver badge

        Re: Can't resist...

        Cut the branch off under them!

        1. EarthDog

          Re: Can't resist...

          root out the negativity

          1. Swarthy

            Re: Can't resist...

            I'm at my limit!

  2. Anonymous Coward
    Anonymous Coward

    I'd ask if it's rational, but that wouldn't be rational :-p

  3. sreynolds

    But how well does it do?

    Surely the yardstick is on how well does it do on the shit from clay test, no?

    1. Charlie Clark Silver badge

      Re: But how well does it do?

      That, but more worringly is adversarial tampering: adding noise that will successfully distract the AI. Imagine the consequences of this in real life if you can effectively scramble road signs, etc.

    2. Cuddles

      Re: But how well does it do?

      Well it was not much better than 50% reliable at recognising a panda, and this technique makes it worse, but not quite as much worse as a slightly different technique. So it might be OK in a low-pressure situation where you don't mind too much if you get a few gibbons mixed in with your pandas, but still doesn't sound great if you plan on setting it loose in a couple of tons of metal careening around the place mere metres from squishy meatbags.

      1. Paul Crawford Silver badge

        Re: But how well does it do?

        "you don't mind too much if you get a few gibbons mixed in with your pandas"

        A phrase I never expected to hear.

        1. Anonymous Coward
          Anonymous Coward

          Showing my age...

          It's potentially a very funky phrase, though... gibbon half a chance.

  4. Pascal Monett Silver badge

    "a more 'terraced' or 'plateaued' landscape to explore"

    Sure, we're already pretending we have AI, why not throw gardening in the mix ?

    I really would appreciate it if El Reg at least could stop pandering to this marketing illusion. It's not AI, it's a statistical analysis machine.

    In any case, call me when they need mowing the lawn. I'll be sure to ensure that all the data points are cut down to the same height.

    1. vtcodger Silver badge

      Re: "a more 'terraced' or 'plateaued' landscape to explore"

      So, you're suggesting that this is an imaginary solution?

  5. Paul Hovnanian Silver badge

    In the training data ...

    ... and the neural net, it sort of makes sense. Instead of representing each point in solution space as a magnitude, it is now a vector (magnitude plus direction) to the next most probable step in that space.

    Once you hit the network with real data, it saves time (processor steps) by proceeding along the most probable path.

  6. Throatwarbler Mangrove Silver badge

    Panda identification ...

    ... apparently not as black-and-white as one might expect.

  7. Il'Geller

    “… correctly labelled by the object recognition algorithm with a 57.7 per cent confidence level, was modified with noise - making the still-very-clearly-a-panda appear to the algorithm as a gibbon with a worrying 93.3 per cent confidence.”

    Great! I no longer need to say that the only innovation — in the past 70 years — is text annotations. And insist that everything depends on textual search.

POST COMMENT House rules

Not a member of The Register? Create a new account here.

  • Enter your comment

  • Add an icon

Anonymous cowards cannot choose their icon

Other stories you might like