One Model To Learn Them All · HackerPapers discussions

One Model To Learn Them All — HN submission

Wed, 30 Oct 2024 19:14:23 GMT

One Model to Learn Them All

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

One Model To Learn Them All — mentioned by phowon

Wed, 27 Mar 2019 19:16:25 GMT

The success of Transformers aside, I'm not sure you should be relying on model titles for anything, lest we forget papers like "One Model To Learn Them All" [1]. [1] https://arxiv.org/abs/1706.05137

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

One Model To Learn Them All — HN submission

Sun, 10 Jun 2018 22:59:35 GMT

One Model to learn them all (2017)

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

One Model To Learn Them All — mentioned by cs702

Sat, 03 Feb 2018 16:22:47 GMT

The author is Douglas Hofstadter, author of Gödel, Escher, Bach.[a] In this article, he shows with concrete examples how Google Translate falls short, and then offers two criticisms: * Feeding more data to current models won't bring them any closer to understanding, since understanding involves having ideas (including ideas about the state of the world), and this lack of ideas is the root of all the problems for machine translation today. His examples are powerful evidence of this limitation of current state-of-the-art machine translation systems. * Current machine translation systems make no…

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

One Model To Learn Them All — mentioned by indescions_2017

Wed, 22 Nov 2017 14:36:46 GMT

A nice complement to this is Andrej Karpathy's Software 2.0 essay. Namely that it is getting much easier to collect the data. Than it ever will be to write the complex programs which can anticipate every case. Software 2.0 https://medium.com/@karpathy/software-2-0-a64152b37c35 One Model To Learn Them All https://arxiv.org/abs/1706.05137

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

One Model To Learn Them All — mentioned by espadrine

Tue, 18 Jul 2017 22:08:32 GMT

> we need better ways of composing this `deep` abstractions Tensor2Tensor[0], from the Google Brain team, has some strong recent results in that direction. Related paper: "One Model To Learn Them All"[1]. [0]: https://github.com/tensorflow/tensor2tensor [1]: https://arxiv.org/pdf/1706.05137.pdf

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

One Model To Learn Them All — HN submission

Sat, 01 Jul 2017 10:34:26 GMT

One Model to Learn Them All

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

One Model To Learn Them All — HN submission

Thu, 29 Jun 2017 17:35:26 GMT

One Model to Learn Them All

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

One Model To Learn Them All — HN submission

Fri, 23 Jun 2017 15:55:33 GMT

Deep Learning: One Model to Learn Them All

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

One Model To Learn Them All — HN submission

Thu, 22 Jun 2017 22:32:34 GMT

One model to learn them all

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

One Model To Learn Them All — HN submission

Wed, 21 Jun 2017 07:18:43 GMT

One Model to Learn Them All

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

One Model To Learn Them All — HN submission

Tue, 20 Jun 2017 10:36:54 GMT

One Model to Learn Them All

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

One Model To Learn Them All — mentioned by bdcs

Mon, 19 Jun 2017 21:48:03 GMT

One Model To Learn Them All Deep learning yields great results across many fields, from speech recognition, image classification, to translation. But for each problem, getting a deep model to work well involves research into the architecture and a long period of tuning. We present a single model that yields good results on a number of problems spanning multiple domains. In particular, this single model is trained concurrently on ImageNet, multiple translation tasks, image captioning (COCO dataset), a speech recognition corpus, and an English parsing task. Our model architecture incorporates…

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

One Model To Learn Them All — HN submission

Mon, 19 Jun 2017 20:32:31 GMT

One Model to Learn Them All

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

One Model To Learn Them All — HN submission

Mon, 19 Jun 2017 10:38:13 GMT

One Model to Learn Them All

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