645      阿裏雲  技術社區[雲棲]

手把手教你使用TensorFlow生成對抗樣本 | 附源碼

https://yq.aliyun.com/cloud

Kaggle

Kaggle

TensorFlow

Jupyter notebook

ImageNetInception v3TF-slim意

import tensorflow as tf
import tensorflow.contrib.slim as slim
import tensorflow.contrib.slim.nets as nets

tf.logging.set_verbosity(tf.logging.ERROR)
sess = tf.InteractiveSession()

tf.Variabletf.placeholder，這是

image = tf.Variable(tf.zeros((299, 299, 3)))

Inception v3

def inception(image, reuse):
    preprocessed = tf.multiply(tf.subtract(tf.expand_dims(image, 0), 0.5), 2.0)
    arg_scope = nets.inception.inception_v3_arg_scope(weight_decay=0.0)
    with slim.arg_scope(arg_scope):
        logits, _ = nets.inception.inception_v3(
            preprocessed, 1001, is_training=False, reuse=reuse)
        logits = logits[:,1:] # ignore background class
        probs = tf.nn.softmax(logits) # probabilities
    return logits, probs

logits, probs = inception(image, reuse=False)

Inception v393.9

import tempfile
from urllib.request import urlretrieve
import tarfile
import os
data_dir = tempfile.mkdtemp()
inception_tarball, _ = urlretrieve(
    'https://download.tensorflow.org/models/inception_v3_2016_08_28.tar.gz')
tarfile.open(inception_tarball, 'r:gz').extractall(data_dir)
restore_vars = [
    var for var in tf.global_variables()
    if var.name.startswith('InceptionV3/')
]
saver = tf.train.Saver(restore_vars)
saver.restore(sess, os.path.join(data_dir, 'inception_v3.ckpt'))

import json
import matplotlib.pyplot as plt
imagenet_json, _ = urlretrieve(
    'https://www.anishathalye.com/media/2017/07/25/imagenet.json')
with open(imagenet_json) as f:
    imagenet_labels = json.load(f)
def classify(img, correct_class=None, target_class=None):
    fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(10, 8))
    fig.sca(ax1)
    p = sess.run(probs, feed_dict={image: img})[0]
    ax1.imshow(img)
    fig.sca(ax1)
    
    topk = list(p.argsort()[-10:][::-1])
    topprobs = p[topk]
    barlist = ax2.bar(range(10), topprobs)
    if target_class in topk:
        barlist[topk.index(target_class)].set_color('r')
    if correct_class in topk:
        barlist[topk.index(correct_class)].set_color('g')
    plt.sca(ax2)
    plt.ylim([0, 1.1])
    plt.xticks(range(10),
               [imagenet_labels[i][:15] for i in topk],
               rotation='vertical')
    fig.subplots_adjust(bottom=0.2)
    plt.show()

import PIL
import numpy as np
img_path, _ = urlretrieve('https://www.anishathalye.com/media/2017/07/25/cat.jpg')
img_class = 281
img = PIL.Image.open(img_path)
big_dim = max(img.width, img.height)
wide = img.width > img.height
new_w = 299 if not wide else int(img.width * 299 / img.height)
new_h = 299 if wide else int(img.height * 299 / img.width)
img = img.resize((new_w, new_h)).crop((0, 0, 299, 299))
img = (np.asarray(img) / 255.0).astype(np.float32)
classify(img, correct_class=img_class)

對抗樣本

XP(y|X)X'logP(y'|X'）被最大化為目標標簽y'ε‖X- X'‖∞≤εX'X

梯度下降來解決基本上訓練網絡本身相同的技術算法很簡單：

X'←X重複以下直到收斂

1. X'←X^+α⋅∇logP(y'|X'）

2. X'←clip(X'

TensorFlow op

x = tf.placeholder(tf.float32, (299, 299, 3))

x_hat = image # our trainable adversarial input
assign_op = tf.assign(x_hat, x)

梯度下降步驟

learning_rate = tf.placeholder(tf.float32, ())
y_hat = tf.placeholder(tf.int32, ())

labels = tf.one_hot(y_hat, 1000)
loss = tf.nn.softmax_cross_entropy_with_logits(logits=logits, labels=[labels])
optim_step = tf.train.GradientDescentOptimizer(
    learning_rate).minimize(loss, var_list=[x_hat])

[01]

epsilon = tf.placeholder(tf.float32, ())

below = x - epsilon
above = x + epsilon
projected = tf.clip_by_value(tf.clip_by_value(x_hat, below, above), 0, 1)
with tf.control_dependencies([projected]):
    project_step = tf.assign(x_hat, projected)

“

demo_epsilon = 2.0/255.0 # a really small perturbation
demo_lr = 1e-1
demo_steps = 100
demo_target = 924 # "guacamole"

# initialization step
sess.run(assign_op, feed_dict={x: img})

# projected gradient descent
for i in range(demo_steps):
    # gradient descent step
    _, loss_value = sess.run(
        [optim_step, loss],
        feed_dict={learning_rate: demo_lr, y_hat: demo_target})
    # project step
    sess.run(project_step, feed_dict={x: img, epsilon: demo_epsilon})
    if (i+1) % 10 == 0:
        print('step %d, loss=%g' % (i+1, loss_value))
    

adv = x_hat.eval() # retrieve the adversarial example

step 10, loss=4.18923
step 20, loss=0.580237
step 30, loss=0.0322334
step 40, loss=0.0209522
step 50, loss=0.0159688
step 60, loss=0.0134457
step 70, loss=0.0117799
step 80, loss=0.0105757
step 90, loss=0.00962179
step 100, loss=0.00886694

“

classify(adv, correct_class=img_class, target_class=demo_target)

的對抗

θ∈[- π/4

θ=π/8

ex_angle = np.pi/8

angle = tf.placeholder(tf.float32, ())
rotated_image = tf.contrib.image.rotate(image, angle)
rotated_example = rotated_image.eval(feed_dict={image: adv, angle: ex_angle})
classify(rotated_example, correct_class=img_class, target_class=demo_target)

如何使TE_t~TlogP(y'|t(X'))‖X- X'‖∞≤ε∇E_t~TlogP(y'|t(X'))與E_t~T∇logP(y'|t(X'))相等樣本

TensorFlow

num_samples = 10
average_loss = 0
for i in range(num_samples):
    rotated = tf.contrib.image.rotate(
        image, tf.random_uniform((), minval=-np.pi/4, maxval=np.pi/4))
    rotated_logits, _ = inception(rotated, reuse=True)
    average_loss += tf.nn.softmax_cross_entropy_with_logits(
        logits=rotated_logits, labels=labels) / num_samples

assign_opproject_stepoptim_step

PGD“

demo_epsilon = 8.0/255.0 # still a pretty small perturbation
demo_lr = 2e-1
demo_steps = 300
demo_target = 924 # "guacamole"

# initialization step
sess.run(assign_op, feed_dict={x: img})

# projected gradient descent
for i in range(demo_steps):
    # gradient descent step
    _, loss_value = sess.run(
        [optim_step, average_loss],
        feed_dict={learning_rate: demo_lr, y_hat: demo_target})
    # project step
    sess.run(project_step, feed_dict={x: img, epsilon: demo_epsilon})
    if (i+1) % 50 == 0:
        print('step %d, loss=%g' % (i+1, loss_value))
    

adv_robust = x_hat.eval() # retrieve the adversarial example

step 50, loss=0.0804289
step 100, loss=0.0270499
step 150, loss=0.00771527
step 200, loss=0.00350717
step 250, loss=0.00656128
step 300, loss=0.00226182

這種對抗圖像被高度信任地歸類為“鱷梨醬”，即使是旋轉的情況下！

rotated_example = rotated_image.eval(feed_dict={image: adv_robust, angle: ex_angle})
classify(rotated_example, correct_class=img_class, target_class=demo_target)

P(y'|x')θ∈[- π/4

thetas = np.linspace(-np.pi/4, np.pi/4, 301)

p_naive = []
p_robust = []
for theta in thetas:
    rotated = rotated_image.eval(feed_dict={image: adv_robust, angle: theta})
    p_robust.append(probs.eval(feed_dict={image: rotated})[0][demo_target])
    
    rotated = rotated_image.eval(feed_dict={image: adv, angle: theta})
    p_naive.append(probs.eval(feed_dict={image: rotated})[0][demo_target])

robust_line, = plt.plot(thetas, p_robust, color='b', linewidth=2, label='robust')
naive_line, = plt.plot(thetas, p_naive, color='r', linewidth=2, label='naive')
plt.ylim([0, 1.05])
plt.xlabel('rotation angle')
plt.ylabel('target class probability')
plt.legend(handles=[robust_line, naive_line], loc='lower right')
plt.show()

Anish AthalyeMIT

https://www.anish.io/

Emailaathalye@mit.edu

Github: https://github.com/anishathalye

@

A Step-by-Step Guide to Synthesizing Adversarial Examples

最後更新：2017-07-28 23:35:47

  上一篇：  中小企業域名選購方略

  下一篇：  智慧醫療給孩子們一個學醫的理由