3 ms·
Is there much in the way of datasets of things that are recognizable but triggering incorrect something-or-other (what is it exactly? Semantic sensations?) Whe
by Lerc 2y ago
Is there much in the way of datasets of things that are recognizable but triggering incorrect something-or-other (what is it exactly? Semantic sensations?)
When we as humans see a face in a cookie or something we quite often spend some time gazing at it noting what it is about the the arrangement of features that triggers the pareidolia. I have wondered if that urge to contemplate the similarity difference is an instinctive response to gather more data for future accuracy.
I guess in a similar vein there would be merit in collecting a data set of all of those quirks that generative AI produces to provide a point of comparison. Notably when most image generators produce artifacts it's because the local image generation is accurate but at odds with the global image. So fingers are beside fingers accurately but there are too many, or a swirl near some hair becomes hair but doesn't actually connect to the top of the head.
- bossyTeacher 2y agoYou are looking at it from your own rationalist point of view. Most humans will apply supernatural, magical, religious or/and spiritual meanings to it. This applies to faces in surfaces as well voices in wind or patterns in noise.
- gwern 2y ago> Is there much in the way of datasets of things that are recognizable but triggering incorrect something-or-other (what is it exactly? Semantic sensations?) There are some datasets which collect 'hard' examples. One I love for the title alone is "When does dough become a bagel? Analyzing the remaining mistakes on ImageNet" https://arxiv.org/abs/2205.04596#google https://arxiv.org/abs/2205.04596#google , Vasudevan et al 2022. The errors can be pretty interesting: https://arxiv.org/pdf/2205.04596#page=18 https://arxiv.org/pdf/2205.04596#page=18 You can often see why a model might make a mistake and that you would too. (Like that first one of a swing - I would think it's some sort of large "tripod" too because I have no idea if I've ever seen a swing like that.)
- dekhn 2y agoThe very idea that some objects have a single, correct label is fundamentally a simplification.
- no_news_is 2y agoIn that example, Ground Truth is a tripod and the Prediction is a swing; that is, you'd be right to think it's a tripod. Of course your statement still stands regardless - the errors are interesting.
- Jerrrrrrry 2y agohttps://en.wikipedia.org/wiki/Adversarial_machine_learning#Adversarial_deep_reinforcement_learning https://en.wikipedia.org/wiki/Adversarial_machine_learning#A...