I disagree. It's difficult to understand and fix bias through #AI algorithms for #DeepLearning These are highly non-linear blackbox models. they amplify biases. many works show that trying to fix superficially is like putting lipstick on a pig. We need fresh thinkinghttps://twitter.com/ylecun/status/1203211859366576128 …
-
Show this thread
-
Prof. Anima Anandkumar Retweeted Prof. Anima Anandkumar
In a recent work we report an intriguing finding to detect bias : assess hardness of different samples for
#DeepLearning model. It's a first step. Using it to fix bias is much harderhttps://twitter.com/animaanandkumar/status/1203090855097057280 …Prof. Anima Anandkumar added,
Prof. Anima Anandkumar @AnimaAnandkumarIntriguing measure of#bias in#DeepLearning We find angular distance as a robust + universal measure of hardness of a training example and corresponds with human ambiguity.@beidichen@animesh_garg@jankautz@NvidiaAI https://twitter.com/Deep__AI/status/1203028047651205120 …Show this thread3 replies 6 retweets 64 likesShow this thread -
More importantly our work shows that
#DeepLearning model is deciding which samples are hard for it to classify and introducing bias. Here I use bias to mean disparate treatment not statistical bias. Model leaves harder example with worse accuracy and more vulnerability to noise1 reply 2 retweets 42 likesShow this thread -
So which
#DeepLearning model you choose changes the bias introduced. We show that older models like Alexnet are worse compared to newer models like resnet. So that's a good thing.@BeidiChen@Anshumali_@animesh_garg@jankautz@NvidiaAI1 reply 1 retweet 38 likesShow this thread -
It's important for our
#AI leaders to acknowledge that#DeepLearning makes it harder to deal with bias. Models themselves introduce bias. It's important for our leaders not to be so dismissive of deep work happening in this area. They should listen and learn from others.5 replies 8 retweets 67 likesShow this thread -
Replying to @AnimaAnandkumar
Isn't Yann talking about social bias, like racial or gender disparity in ML outcomes? The bias you reportt is totally real, but has no reason to equate to recognisable social bias (vs other random bias). Or are you saying that social bias is consistently identified by hardness?
1 reply 0 retweets 0 likes
The two are related certainly, and further research needs to be done to understand it.
-
-
Replying to @AnimaAnandkumar @DrLukeOR
These are approaching
#legaltech considerations fast - think policies and norms.0 replies 0 retweets 1 likeThanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.