ILPD¶
ILPD
¶
Bases: Attack
The ILPD (Intermediate-level Perturbation Decay) Attack.
From the paper: Improving Adversarial Transferability via Intermediate-level Perturbation Decay.
Parameters:
Name | Type | Description | Default |
---|---|---|---|
model
|
Module | AttackModel
|
The model to attack. |
required |
normalize
|
Callable[[Tensor], Tensor] | None
|
A transform to normalize images. |
None
|
device
|
device | None
|
Device to use for tensors. Defaults to cuda if available. |
None
|
eps
|
float
|
The maximum perturbation. Defaults to 8/255. |
8 / 255
|
steps
|
int
|
Number of steps. Defaults to 10. |
10
|
alpha
|
float | None
|
Step size, |
None
|
decay
|
float
|
Decay factor for the momentum term. Defaults to 1.0. |
1.0
|
sigma
|
float
|
Standard deviation for noise. Defaults to 0.05. |
0.05
|
feature_layer_cfg
|
str
|
Name of the feature layer to attack. If not provided, tries to infer from built-in config based on the model name. Defaults to "" |
''
|
clip_min
|
float
|
Minimum value for clipping. Defaults to 0.0. |
0.0
|
clip_max
|
float
|
Maximum value for clipping. Defaults to 1.0. |
1.0
|
Source code in torchattack/ilpd.py
11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 |
|
forward(x, y)
¶
Perform ILPD on a batch of images.
Parameters:
Name | Type | Description | Default |
---|---|---|---|
x
|
Tensor
|
A batch of images. Shape: (N, C, H, W). |
required |
y
|
Tensor
|
A batch of labels. Shape: (N). |
required |
Returns:
Type | Description |
---|---|
Tensor
|
The perturbed images if successful. Shape: (N, C, H, W). |