ExGAN: Adversarial Generation of Extreme Samples · HackerPapers discussions

ExGAN: Adversarial Generation of Extreme Samples — mentioned by siddhartb_

Tue, 22 Sep 2020 14:13:12 GMT

Existing GAN based approaches excel at generating realistic samples, but seek to generate typical samples, rather than extreme samples. We propose ExGAN to generate realistic and extreme samples. ExGAN allows the user to specify both the desired extremeness measure, as well as the desired extremeness probability they wish to sample at. Generating increasingly extreme examples can be done in constant time (with respect to the extremeness probability), as opposed to the exponential time required by the baseline. Paper: https://arxiv.org/abs/2009.08454

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

ExGAN: Adversarial Generation of Extreme Samples — HN submission

Tue, 22 Sep 2020 10:54:45 GMT

Show HN: Adversarial Generation of Extreme Samples

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

ExGAN: Adversarial Generation of Extreme Samples — HN submission

Fri, 18 Sep 2020 12:42:33 GMT

ExGAN: Adversarial Generation of Extreme Samples

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