Putting An End to End-to-End: Gradient-Isolated Learning of Representations · HackerPapers discussions

Putting An End to End-to-End: Gradient-Isolated Learning of Representations — mentioned by LudwigNagasena

Thu, 12 Jan 2023 15:51:53 GMT

The question was "If it's not new or novel, why aren't people using it?". For example, take a look at this paper: https://arxiv.org/abs/1905.11786. It was published 3 years ago, it also does parallel layer-wise optimization, it even talks about being inspired by biology, though the objective function is different from Hinton's. Why aren't people using it? Because it performs worse. It is that simple. Is it an interesting area to explore? Probably. There are millions of interesting areas to explore. It doesn't mean it is worth using, at least yet.

Paper and discussion history. A mention is not an endorsement.

Putting An End to End-to-End: Gradient-Isolated Learning of Representations — mentioned by 1_over_n

Fri, 06 Mar 2020 02:34:22 GMT

similar claim made in another paper posted a few days ago https://arxiv.org/abs/1905.11786

Paper and discussion history. A mention is not an endorsement.

Putting An End to End-to-End: Gradient-Isolated Learning of Representations — HN submission

Tue, 03 Mar 2020 12:52:08 GMT

Putting an End to End-to-End: Gradient-Isolated Learning of Representations

2 points; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.