News for hackers. Stuff that matters.
Re:Where we see shapes, AI sees textures by Majromax on Monday July 08, 2019 @03:10PM (#20385773)
> Texture is a more informative feature than shape in the datasets used.

But the insight is that although both features are informative, classical training approaches result in networks that over-fit on texture information and ignore shape information.

Table 2 of the discussed paper finds that joint training with "Stylized ImageNet" (IN images after a style transfer) and classic IN followed by a fine-tuning pass on IN results in improved classification accuracy over IN training alone.

Re:Where we see shapes, AI sees textures by skgoa on Tuesday July 09, 2019 @04:32AM (#20389912)
Which proves exactly what the top level comment claimed: the distribution of examples in the dataset was incorrect for the learning task.
Re:Where we see shapes, AI sees textures by yarg on Tuesday July 09, 2019 @08:16AM (#20391001)
Not necessarily. Textures are high frequency low abstraction data forms, shapes are low frequency high abstraction data forms.

The former are easier to learn for multiple reasons: The nature of higher frequency data means that you have more samples.

(There are far more furry textures in a picture of a cat than there are toes, paws, legs etc.)

Low abstraction data forms are also learnt in earlier layers of the network - more abstract forms are resolved through the understanding of how less abstract forms group together, making the low abstraction forms a prerequisite to more in depth learning.