How we built the Waifu Vending Machine

Re:How we built the Waifu Vending Machine by ipsum2 on Wednesday July 24, 2019 @12:46AM (#20512416)

The ImageNet dataset, which is probably the most commonly used dataset (besides MNIST), is also "copyright-infringement on a gross scale". http://image-net.org/about-overview

[ Parent | on HN ]

Re:How we built the Waifu Vending Machine by tofof on Wednesday July 24, 2019 @12:53AM (#20512450)

Yes, it is.

The fact that other datasets are also massively infringing is evidence of a severe problem for the field, not an exoneration.

[ Parent | on HN ]

Re:How we built the Waifu Vending Machine by DuskStar on Wednesday July 24, 2019 @03:43AM (#20513081)

I would personally describe it as a massive failure of copyright law, not of the datasets.

[ Parent | on HN ]