A Metric Learning Reality Check · HackerPapers discussions

A Metric Learning Reality Check — mentioned by igorkraw

Thu, 01 Dec 2022 22:38:08 GMT

Ben Recht actually has a line of work showing that we aren't over fitting the validation/test set for now (amazingly...). What I mean is, by chasing higher and higher SotA with more and more money and compute, whole fields can go "improving" only for papers like https://arxiv.org/abs/2003.08505 or "Implementation matters in deep RL" to come out and show that what's going on is different from the literature consensus. The standards for showing improvement are low, while standards for négative résultats are high (I'm a bit biased because I have a rejected paper trying to show empirically some…

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

A Metric Learning Reality Check — mentioned by Eridrus

Wed, 10 Jun 2020 14:13:21 GMT

Another paper on the related topic of metric learning arguing that metric learning hasn't actually made any progress and is piggybacking on progress elsewhere: https://arxiv.org/abs/2003.08505

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

A Metric Learning Reality Check — mentioned by chriskanan

Wed, 27 May 2020 12:27:06 GMT

AI is having some reproducibility issues, especially reinforcement learning: https://www.wired.com/story/artificial-intelligence-confront... In other areas of machine learning, things have gotten a lot better. A decade ago people rarely released their code or trained models. I did this for an early feature learning paper in 2010, which led to it getting a lot of citations. What I'm more worried about is the lack of science in machine learning papers. The code reproduces the result, but the reasons for the efficacy given in the paper are spurious. I like this recent paper that pokes this issue…

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

A Metric Learning Reality Check — HN submission

Mon, 11 May 2020 20:02:44 GMT

A Metric Learning Reality Check

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

A Metric Learning Reality Check — HN submission

Mon, 11 May 2020 03:57:56 GMT

A Deep Metric Learning Reality Check

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

A Metric Learning Reality Check — HN submission

Sun, 10 May 2020 17:32:05 GMT

A Metric Learning Reality Check

2 points; 1 discussion reply on HN. Paper and source history. A submission is not an endorsement.