That was pretty funny, but I think the 1 in 1,000,000 failure rate mention was about security, not execution, so a false negative in this situation probably helps the point. @tbodt
is still betting against the 1 in 1,000,000 in his comment though π
It may be true that 1 in 1,000,000 people can unlock your phone, but what about carefully constructed adversarial examples? If you do research on breaking machine learning, see if you can get the lab to pay for an iPhone X to experiment on.
Also, 1 in 1,000,000 means there's on the order of 10,000 people out there who can unlock your phone.
For further actions, you may consider blocking this person and/or reporting abuse
We're a place where coders share, stay up-to-date and grow their careers.
That was pretty funny, but I think the 1 in 1,000,000 failure rate mention was about security, not execution, so a false negative in this situation probably helps the point. @tbodt is still betting against the 1 in 1,000,000 in his comment though π
It may be true that 1 in 1,000,000 people can unlock your phone, but what about carefully constructed adversarial examples? If you do research on breaking machine learning, see if you can get the lab to pay for an iPhone X to experiment on.
Also, 1 in 1,000,000 means there's on the order of 10,000 people out there who can unlock your phone.