Corporations are slowly integrating this technology and sell these off as features on newer generation phones like face recognition or finger print unlocking without the average Joe not knowing the potential threat it might pose.
Clean data is essential to good results in AI and machine learning, but data can become biased and less accurate at multiple stages in its lifetime—from moment it is generated all the way through to when it is processed—and it can happen in ways that are not always obvious and often difficult to discern https://semiengineering.com/where-data-gets-biased/
Your not wrong, but I think this is worth taking a second to clarify.
See, when I think of "clean" data, I think of a set of crisp, noiseless images, or text without typos, or DNA sequences w/o sequencing errors or low-quality reads.
For AI purposes, you want a set of representative noise... w/o it, you'll be building a very brittle model. The key is, as you've mentioned, to assure that the method of data collection/generation/processing is either (1) unbiased throughout or (2) is done using various methods ideally performed by different entities.
So, there is certainly a distinction between bias and noise (the opposite of which I would probably call "cleanliness").
I don't see why this was downvoted. We really are heading towards exactly such a future, in which the only psuedo-chance of recapturing privacy will be through surgeries that modify biometric indicators.
5 comments
[ 2.9 ms ] story [ 20.8 ms ] threadYour not wrong, but I think this is worth taking a second to clarify.
See, when I think of "clean" data, I think of a set of crisp, noiseless images, or text without typos, or DNA sequences w/o sequencing errors or low-quality reads.
For AI purposes, you want a set of representative noise... w/o it, you'll be building a very brittle model. The key is, as you've mentioned, to assure that the method of data collection/generation/processing is either (1) unbiased throughout or (2) is done using various methods ideally performed by different entities.
So, there is certainly a distinction between bias and noise (the opposite of which I would probably call "cleanliness").