Skip to content

Commit e1dcf9c

Browse files
authored
fix: major formatting changes (#62)
resolves: #61
1 parent 4e0ab01 commit e1dcf9c

10 files changed

+568
-557
lines changed

2023/ML01_2023-Adversarial_Attack.md

Lines changed: 50 additions & 51 deletions
Original file line numberDiff line numberDiff line change
@@ -4,77 +4,76 @@ type: documentation
44
altfooter: true
55
level: 4
66
auto-migrated: 0
7-
pitch:
87
document: OWASP Machine Learning Security Top Ten 2023
98
year: 2023
109
order: 1
1110
title: ML01:2023 Adversarial Attack
1211
lang: en
13-
author:
14-
contributors:
15-
tags: OWASP Machine Learning Security Top Ten 2023, Top Ten, ML01:2023, mltop10, mlsectop10
12+
tags:
13+
[
14+
OWASP Machine Learning Security Top Ten 2023,
15+
Top Ten,
16+
ML01:2023,
17+
mltop10,
18+
mlsectop10,
19+
]
1620
exploitability: 5
17-
prevalence:
1821
detectability: 3
1922
technical: 5
20-
redirect_from:
2123
---
2224

23-
RISK Chart for Scenario One:
25+
## Description
2426

25-
| Threat agents/Attack vectors | Security Weakness | Impact |
26-
| ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | :-----------------------------------------------------------------------------------------------------------------------: | :-----------------------------------------------------------------------------------------------------: |
27-
| Exploitability: 5 (Easy to exploit) ML Application Specific: 4 ML Operations Specific: 3 | Detectability: 3 (The adversarial image may not be noticeable to the naked eye, making it difficult to detect the attack) | Technical: 5 (The attack requires technical knowledge of deep learning and image processing techniques) |
28-
| Threat Agent: Attacker with knowledge of deep learning and image processing techniques Attack Vector: Deliberately crafted adversarial image that is similar to a legitimate image | Vulnerability in the deep learning model's ability to classify images accurately | Misclassification of the image, leading to security bypass or harm to the system |
27+
Adversarial attacks are a type of attack in which an attacker deliberately
28+
alters input data to mislead the model.
2929

30-
It is important to note that this chart is only a sample based on
31-
scenario below, and the actual risk assessment will depend on the
32-
specific circumstances of each machine learning system.
30+
## How to Prevent
31+
32+
**Adversarial training:** One approach to defending against adversarial attacks
33+
is to train the model on adversarial examples. This can help the model become
34+
more robust to attacks and reduce its susceptibility to being misled.
3335

34-
**Description**:
35-
Adversarial attacks are a type of attack in which an attacker
36-
deliberately alters input data to mislead the model.
36+
**Robust models:** Another approach is to use models that are designed to be
37+
robust against adversarial attacks, such as adversarial training or models that
38+
incorporate defense mechanisms.
3739

38-
**Example Attack Scenario:**
40+
**Input validation:** Input validation is another important defense mechanism
41+
that can be used to detect and prevent adversarial attacks. This involves
42+
checking the input data for anomalies, such as unexpected values or patterns,
43+
and rejecting inputs that are likely to be malicious.
3944

40-
Scenario 1: Image classification
45+
## Risk Factors
4146

42-
A deep learning model is trained to classify images into different
43-
categories, such as dogs and cats. An attacker creates an adversarial
44-
image that is very similar to a legitimate image of a cat, but with
45-
small, carefully crafted perturbations that cause the model to
46-
misclassify it as a dog. When the model is deployed in a real-world
47-
setting, the attacker can use the adversarial image to bypass security
48-
measures or cause harm to the system.
47+
| Threat Agents/Attack Vectors | Security Weakness | Impact |
48+
| --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | :--------------------------------------------------------------------------------------------------------------------------------------------: | :---------------------------------------------------------------------------------------------------------------------------: |
49+
| Exploitability: 5 (Easy) <br><br> _ML Application Specific: 4_ <br> _ML Operations Specific: 3_ | Detectability: 3 (Moderate) <br><br> _The adversarial image may not be noticeable to the naked eye, making it difficult to detect the attack._ | Technical: 5 (Difficult) <br><br> _The attack requires technical knowledge of deep learning and image processing techniques._ |
50+
| Threat Agent: Attacker with knowledge of deep learning and image processing techniques. <br><br> Attack Vector: Deliberately crafted adversarial image that is similar to a legitimate image. | Vulnerability in the deep learning model's ability to classify images accurately. | Misclassification of the image, leading to security bypass or harm to the system. |
4951

50-
Scenario 2: Network intrusion detection
52+
It is important to note that this chart is only a sample based on
53+
[the scenario below](#scenario1) only. The actual risk assessment will depend on
54+
the specific circumstances of each machine learning system.
5155

52-
A deep learning model is trained to detect intrusions in a network. An
53-
attacker creates adversarial network traffic by carefully crafting
54-
packets in such a way that they will evade the model\'s intrusion
55-
detection system. The attacker can manipulate the features of the
56-
network traffic, such as the source IP address, destination IP address,
57-
or payload, in such a way that they are not detected by the intrusion
58-
detection system. For example, the attacker may hide their source IP
59-
address behind a proxy server or encrypt the payload of their network
60-
traffic. This type of attack can have serious consequences, as it can
61-
lead to data theft, system compromise, or other forms of damage.
56+
## Example Attack Scenarios
6257

63-
**How to Prevent:**
58+
### Scenario \#1: Image classification {#scenario1}
6459

65-
1. Adversarial training: One approach to defending against adversarial
66-
attacks is to train the model on adversarial examples. This can help
67-
the model become more robust to attacks and reduce its
68-
susceptibility to being misled.
60+
A deep learning model is trained to classify images into different categories,
61+
such as dogs and cats. An attacker creates an adversarial image that is very
62+
similar to a legitimate image of a cat, but with small, carefully crafted
63+
perturbations that cause the model to misclassify it as a dog. When the model is
64+
deployed in a real-world setting, the attacker can use the adversarial image to
65+
bypass security measures or cause harm to the system.
6966

70-
2. Robust models: Another approach is to use models that are designed
71-
to be robust against adversarial attacks, such as adversarial
72-
training or models that incorporate defense mechanisms.
67+
### Scenario \#2: Network intrusion detection
7368

74-
3. Input validation: Input validation is another important defense
75-
mechanism that can be used to detect and prevent adversarial
76-
attacks. This involves checking the input data for anomalies, such
77-
as unexpected values or patterns, and rejecting inputs that are
78-
likely to be malicious.
69+
A deep learning model is trained to detect intrusions in a network. An attacker
70+
creates adversarial network traffic by carefully crafting packets in such a way
71+
that they will evade the model\'s intrusion detection system. The attacker can
72+
manipulate the features of the network traffic, such as the source IP address,
73+
destination IP address, or payload, in such a way that they are not detected by
74+
the intrusion detection system. For example, the attacker may hide their source
75+
IP address behind a proxy server or encrypt the payload of their network
76+
traffic. This type of attack can have serious consequences, as it can lead to
77+
data theft, system compromise, or other forms of damage.
7978

80-
**References:**
79+
## References

2023/ML02_2023-Data_Poisoning_Attack.md

Lines changed: 63 additions & 61 deletions
Original file line numberDiff line numberDiff line change
@@ -4,92 +4,94 @@ type: documentation
44
altfooter: true
55
level: 4
66
auto-migrated: 0
7-
pitch:
87
document: OWASP Machine Learning Security Top Ten 2023
98
year: 2023
109
order: 2
1110
title: ML02:2023 Data Poisoning Attack
1211
lang: en
13-
author:
14-
contributors:
15-
tags: OWASP Machine Learning Security Top Ten 2023, Top Ten, ML02:2023, mltop10, mlsectop10
12+
tags:
13+
[
14+
OWASP Machine Learning Security Top Ten 2023,
15+
Top Ten,
16+
ML02:2023,
17+
mltop10,
18+
mlsectop10,
19+
]
1620
exploitability: 3
17-
prevalence:
1821
detectability: 2
1922
technical: 4
20-
redirect_from:
2123
---
2224

23-
| Threat agents/Attack vectors | Security Weakness | Impact |
24-
| :--------------------------------------------------------------------------------------------------------------------------------------------------------------: | :-----------------------------------------------------------------------: | :------------------------------------------------------------------------------------------------------------------------------------: |
25-
| Exploitability: 3 (Medium to exploit)<br>ML Application Specific: 4 <br>ML Operations Specific: 3 | Detectability: 2<br>(Limited) | Technical: 4<br> |
26-
| Threat Agent: Attacker who has access to the training data used for the model.<br>Attack Vector: The attacker injects malicious data into the training data set. | Lack of data validation and insufficient monitoring of the training data. | The model will make incorrect predictions based on the poisoned data, leading to false decisions and potentially serious consequences. |
25+
## Description
2726

28-
It is important to note that this chart is only a sample based on
29-
scenario below, and the actual risk assessment will depend on the
30-
specific circumstances of each machine learning system.
27+
Data poisoning attacks occur when an attacker manipulates the training data to
28+
cause the model to behave in an undesirable way.
3129

32-
**Description:**
30+
## How to Prevent
3331

34-
Data poisoning attacks occur when an attacker manipulates the training
35-
data to cause the model to behave in an undesirable way.
32+
**Data validation and verification:** Ensure that the training data is
33+
thoroughly validated and verified before it is used to train the model. This can
34+
be done by implementing data validation checks and employing multiple data
35+
labelers to validate the accuracy of the data labeling.
3636

37-
**Example Attack Scenario:**
37+
**Secure data storage:** Store the training data in a secure manner, such as
38+
using encryption, secure data transfer protocols, and firewalls.
3839

39-
Scenario 1: Training a spam classifier
40+
**Data separation:** Separate the training data from the production data to
41+
reduce the risk of compromising the training data.
4042

41-
An attacker poisons the training data for a deep learning model that
42-
classifies emails as spam or not spam. The attacker executed this attack
43-
by injecting the maliciously labeled spam emails into the training data
44-
set. This could be done by compromising the data storage system, for
45-
example by hacking into the network or exploiting a vulnerability in the
46-
data storage software. The attacker could also manipulate the data
47-
labeling process, such as by falsifying the labeling of the emails or by
48-
bribing the data labelers to provide incorrect labels.
43+
**Access control:** Implement access controls to limit who can access the
44+
training data and when they can access it.
4945

50-
Scenario 2: Training a network traffic classification system
46+
**Monitoring and auditing:** Regularly monitor the training data for any
47+
anomalies and conduct audits to detect any data tampering.
5148

52-
An attacker poisons the training data for a deep learning model that is
53-
used to classify network traffic into different categories, such as
54-
email, web browsing, and video streaming. They introduce a large number
55-
of examples of network traffic that are incorrectly labeled as a
56-
different type of traffic, causing the model to be trained to classify
57-
this traffic as the incorrect category. As a result, the model may be
58-
trained to make incorrect traffic classifications when the model is
59-
deployed, potentially leading to misallocation of network resources or
60-
degradation of network performance.
49+
**Model validation:** Validate the model using a separate validation set that
50+
has not been used during training. This can help to detect any data poisoning
51+
attacks that may have affected the training data.
6152

62-
**How to Prevent:**
53+
**Model ensembles:** Train multiple models using different subsets of the
54+
training data and use an ensemble of these models to make predictions. This can
55+
reduce the impact of data poisoning attacks as the attacker would need to
56+
compromise multiple models to achieve their goals.
6357

64-
Data validation and verification: Ensure that the training data is
65-
thoroughly validated and verified before it is used to train the model.
66-
This can be done by implementing data validation checks and employing
67-
multiple data labelers to validate the accuracy of the data labeling.
58+
**Anomaly detection:** Use anomaly detection techniques to detect any abnormal
59+
behavior in the training data, such as sudden changes in the data distribution
60+
or data labeling. These techniques can be used to detect data poisoning attacks
61+
early on.
6862

69-
Secure data storage: Store the training data in a secure manner, such as
70-
using encryption, secure data transfer protocols, and firewalls.
63+
## Risk Factors
7164

72-
Data separation: Separate the training data from the production data to
73-
reduce the risk of compromising the training data.
65+
| Threat Agents/Attack Vectors | Security Weakness | Impact |
66+
| :--------------------------------------------------------------------------------------------------------------------------------------------------------------------: | :-----------------------------------------------------------------------: | :------------------------------------------------------------------------------------------------------------------------------------: |
67+
| Exploitability: 3 (Moderate) <br><br> _ML Application Specific: 4_ <br> _ML Operations Specific: 3_ | Detectability: 2 (Difficult) | Technical: 4 (Moderate) |
68+
| Threat Agent: Attacker who has access to the training data used for the model. <br><br> Attack Vector: The attacker injects malicious data into the training data set. | Lack of data validation and insufficient monitoring of the training data. | The model will make incorrect predictions based on the poisoned data, leading to false decisions and potentially serious consequences. |
7469

75-
Access control: Implement access controls to limit who can access the
76-
training data and when they can access it.
70+
It is important to note that this chart is only a sample based on
71+
[the scenario below](#scenario1) only. The actual risk assessment will depend on
72+
the specific circumstances of each machine learning system.
7773

78-
Monitoring and auditing: Regularly monitor the training data for any
79-
anomalies and conduct audits to detect any data tampering.
74+
## Example Attack Scenarios
75+
76+
### Scenario \#1: Training a spam classifier {#scenario1}
8077

81-
Model validation: Validate the model using a separate validation set
82-
that has not been used during training. This can help to detect any data
83-
poisoning attacks that may have affected the training data.
78+
An attacker poisons the training data for a deep learning model that classifies
79+
emails as spam or not spam. The attacker executed this attack by injecting the
80+
maliciously labeled spam emails into the training data set. This could be done
81+
by compromising the data storage system, for example by hacking into the network
82+
or exploiting a vulnerability in the data storage software. The attacker could
83+
also manipulate the data labeling process, such as by falsifying the labeling of
84+
the emails or by bribing the data labelers to provide incorrect labels.
8485

85-
Model ensembles: Train multiple models using different subsets of the
86-
training data and use an ensemble of these models to make predictions.
87-
This can reduce the impact of data poisoning attacks as the attacker
88-
would need to compromise multiple models to achieve their goals.
86+
### Scenario \#2: Training a network traffic classification system
8987

90-
Anomaly detection: Use anomaly detection techniques to detect any
91-
abnormal behavior in the training data, such as sudden changes in the
92-
data distribution or data labeling. These techniques can be used to
93-
detect data poisoning attacks early on.
88+
An attacker poisons the training data for a deep learning model that is used to
89+
classify network traffic into different categories, such as email, web browsing,
90+
and video streaming. They introduce a large number of examples of network
91+
traffic that are incorrectly labeled as a different type of traffic, causing the
92+
model to be trained to classify this traffic as the incorrect category. As a
93+
result, the model may be trained to make incorrect traffic classifications when
94+
the model is deployed, potentially leading to misallocation of network resources
95+
or degradation of network performance.
9496

95-
**References:**
97+
## References

0 commit comments

Comments
 (0)