557622faca0328c0303b824006c5fe11cc87cca1,art/attacks/virtual_adversarial.py,VirtualAdversarialMethod,generate,#VirtualAdversarialMethod#Any#,33

Before Change

            e = np.random.randn(*dims)
            for _ in range(self.max_iter):
                d = self.finite_diff * self._normalize(d)
                e = self.finite_diff * self._normalize(e)
                preds_new = self.classifier.predict(np.stack((val + d, val + e)))

                // Compute KL divergence between logits
                from scipy.stats import entropy
                kl_div1 = entropy(preds[ind], preds_new[0])
                kl_div2 = entropy(preds[ind], preds_new[1])
                d = (kl_div1 - kl_div2) / np.abs(d - e)

            // Apply perturbation and clip
            val = np.clip(val + self.eps * self._normalize(d), clip_min, clip_max)
            x_adv[ind] = val

After Change

                
                // TODO remove for loop
                d_new = d
                for i in range(*dims):
                    d[i] += self.finite_diff
                    preds_new = self.classifier.predict((val + d)[None,...], logits=False)
                    kl_div2 = entropy(preds[ind], preds_new[0])                    
                    d_new[i] = (kl_div2-kl_div1)/self.finite_diff
                    d[i] -= self.finite_diff
                d = d_new

            // Apply perturbation and clip
            val = np.clip(val + self.eps * self._normalize(d), clip_min, clip_max)

Instances