adversarial_training_methods
adversarial_training_methods copied to clipboard
enricivi
Implementation of the methods proposed in **Adversarial Training Methods for Semi-Supervised Text Classification** on IMDB dataset (without pre-training)
In the paper, author said "Although we also in general cannot analytically calculate the virtual adversarial loss, Miyato et al. (2016) proposed to calculate the approximated Eq.(3) efficiently with backpropagation....
Hi, I have some questions about the gradient in VAT. The functrion `get_v_adv_loss(self, ul_batch, p_mult, power_iterations=1)` in `paper_network.ipynb` has one statement: ``` gradient = tf.gradients(kl, [d], aggregation_method=tf.AggregationMethod.EXPERIMENTAL_ACCUMULATE_N)[0] ``` Is this...
hello, I have some questions about dropout in VAT. If I use dropout in VAT, the output distribution will change even without perturbation. thanks!
