U tweetove putem weba ili aplikacija drugih proizvođača možete dodati podatke o lokaciji, kao što su grad ili točna lokacija. Povijest lokacija tweetova uvijek možete izbrisati. Saznajte više
Links to the mentioned papers. MixMatch: https://arxiv.org/abs/1905.02249 Unsupervised Data Augmentation: https://arxiv.org/abs/1904.12848
Key idea in these two papers is to ensure prediction(x) = prediction(x + noise) , where x is an unlabeled example. People have tried all kind of noise, e.g., Gaussian noise, adversarial noise etc. But it looks like data augmentation noise is the real winner.pic.twitter.com/ClUOqbvEsj
To add to Vincent's point above, new findings also include: 1. The method is general (works well for images & texts). 2. The method works well on top of transfer learning (e.g., BERT). You can find these results in Unsupervised Data Augmentation paper: https://arxiv.org/abs/1904.12848
It's quite similar to how @citnaj stabilized video colorization in DeOldify
Indeed! And you too- you wrote the augmentation code!
Unexpected results. Naively, one might expect that directly perturbing the model parameters instead of the data for a unsupervised consistency loss might give similar results. But clearly Not!
Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.
