All Projects ā†’ eth-sri ā†’ Diffai

eth-sri / Diffai

Licence: mit
A certifiable defense against adversarial examples by training neural networks to be provably robust

Programming Languages

python
139335 projects - #7 most used programming language

Projects that are alternatives of or similar to Diffai

Free adv train
Official TensorFlow Implementation of Adversarial Training for Free! which trains robust models at no extra cost compared to natural training.
Stars: āœ­ 127 (-25.73%)
Mutual labels:  attack, robust
Wreckuests
Yet another one hard-hitting tool to run HTTP stress tests šŸŒŒ
Stars: āœ­ 137 (-19.88%)
Mutual labels:  attack
Server Configs Nginx
Nginx HTTP server boilerplate configs
Stars: āœ­ 9,457 (+5430.41%)
Mutual labels:  robust
Nailgun
Nailgun attack on ARM devices.
Stars: āœ­ 114 (-33.33%)
Mutual labels:  attack
Deautherdroid
Additional android app for SpaceHunn's ESP8266 DeAuther.
Stars: āœ­ 93 (-45.61%)
Mutual labels:  attack
Wifi Spam
āœ‰ļøšŸ“” Spam thousands of WiFi access points with custom SSIDs
Stars: āœ­ 92 (-46.2%)
Mutual labels:  attack
Pywsus
Standalone implementation of a part of the WSUS spec. Built for offensive security purposes.
Stars: āœ­ 84 (-50.88%)
Mutual labels:  attack
Instahack
Best Tool For instagram bruteforce hacking Tool By EvilDevil
Stars: āœ­ 139 (-18.71%)
Mutual labels:  attack
Kube Aws Iam Controller
Distribute different AWS IAM credentials to different pods in Kubernetes via secrets.
Stars: āœ­ 137 (-19.88%)
Mutual labels:  robust
Misp Maltego
Set of Maltego transforms to inferface with a MISP Threat Sharing instance, and also to explore the whole MITRE ATT&CK dataset.
Stars: āœ­ 112 (-34.5%)
Mutual labels:  attack
Conserve
[beta] robust portable backup tool in Rust
Stars: āœ­ 107 (-37.43%)
Mutual labels:  robust
Nonce Disrespect
Nonce-Disrespecting Adversaries: Practical Forgery Attacks on GCM in TLS
Stars: āœ­ 103 (-39.77%)
Mutual labels:  attack
Dasynq
Thread-safe cross-platform event loop library in C++
Stars: āœ­ 127 (-25.73%)
Mutual labels:  robust
Statsexpressions
Expressions and dataframes with statistical details šŸ“‰ šŸ“œšŸ”£āœ…
Stars: āœ­ 144 (-15.79%)
Mutual labels:  robust
Wifiphisher
Wifiphisher is a rogue Access Point framework for conducting red team engagements or Wi-Fi security testing. Using Wifiphisher, penetration testers can easily achieve a man-in-the-middle position against wireless clients by performing targeted Wi-Fi association attacks. Wifiphisher can be further used to mount victim-customized web phishing attacks against the connected clients in order to capture credentials (e.g. from third party login pages or WPA/WPA2 Pre-Shared Keys) or infect the victim stations with malwares.
Stars: āœ­ 10,333 (+5942.69%)
Mutual labels:  attack
Pax
šŸ’€ šŸ”“ CLI tool for PKCS7 padding oracle attacks
Stars: āœ­ 90 (-47.37%)
Mutual labels:  attack
Jlsca
Side-channel toolkit in Julia
Stars: āœ­ 114 (-33.33%)
Mutual labels:  attack
Atomicpurpleteam
Atomic Purple Team Framework and Lifecycle
Stars: āœ­ 155 (-9.36%)
Mutual labels:  attack
Cc Attack
Using Socks4/5 proxy to make a multithreading Http-flood/Https-flood (cc) attack.
Stars: āœ­ 145 (-15.2%)
Mutual labels:  attack
Confusable homoglyphs
Ļ²Īæļ½Žļ½†ļ½•Ń•Š°ļ½‚ļ½ŒŠµļ¼æŅ»Īæļ½Š¾É”ļ½Œļ½™ļ½ļ½ˆļ½“
Stars: āœ­ 130 (-23.98%)
Mutual labels:  attack

DiffAI v3 portfolio_view

High Level

DiffAI is a system for training neural networks to be provably robust and for proving that they are robust. The system was developed for the 2018 ICML paper and the 2019 ArXiV Paper.

Background

By now, it is well known that otherwise working networks can be tricked by clever attacks. For example Goodfellow et al. demonstrated a network with high classification accuracy which classified one image of a panda correctly, and a seemingly identical attack picture incorrectly. Many defenses against this type of attack have been produced, but very few produce networks for which provably verifying the safety of a prediction is feasible.

Abstract Interpretation is a technique for verifying properties of programs by soundly overapproximating their behavior. When applied to neural networks, an infinite set (a ball) of possible inputs is passed to an approximating "abstract" network to produce a superset of the possible outputs from the actual network. Provided an appropreate representation for these sets, demonstrating that the network classifies everything in the ball correctly becomes a simple task. The method used to represent these sets is the abstract domain, and the specific approximations are the abstract transformers.

In DiffAI, the entire abstract interpretation process is programmed using PyTorch so that it is differentiable and can be run on the GPU, and a loss function is crafted so that low values correspond to inputs which can be proved safe (robust).

Whats New In v3?

  • Abstract Networks: one can now customize the handling of the domains on a per-layer basis.
  • Training DSL: A DSL has been exposed to allow for custom training regimens with complex parameter scheduling.
  • Cross Loss: The box goal now uses the cross entropy style loss by default as suggested by Gowal et al. 2019
  • Conversion to Onyx: We can now export to the onyx format, and can export the abstract network itself to onyx (so that one can run abstract analysis or training using tensorflow for example).

Requirements

python 3.6.7, and virtualenv, torch 0.4.1.

Recommended Setup

$ git clone https://github.com/eth-sri/DiffAI.git
$ cd DiffAI
$ virtualenv pytorch --python python3.6
$ source pytorch/bin/activate
(pytorch) $ pip install -r requirements.txt

Note: you need to activate your virtualenv every time you start a new shell.

Getting Started

DiffAI can be run as a standalone program. To see a list of arguments, type

(pytorch) $ python . --help

At the minimum, DiffAI expects at least one domain to train with and one domain to test with, and a network with which to test. For example, to train with the Box domain, baseline training (Point) and test against the FGSM attack and the ZSwitch domain with a simple feed forward network on the MNIST dataset (default, if none provided), you would type:

(pytorch) $ python . -d "Point()" -d "Box()" -t "PGD()" -t "ZSwitch()" -n ffnn

Unless otherwise specified by "--out", the output is logged to the folder "out/".
In the folder corresponding to the experiment that has been run, one can find the saved configuration options in "config.txt", and a pickled net which is saved every 10 epochs (provided that testing is set to happen every 10th epoch).

To load a saved model, use "--test" as per the example:

(pytorch) $ alias test-diffai="python . -d Point --epochs 1 --dont-write --test-freq 1"
(pytorch) $ test-diffai -t Box --update-test-net-name convBig --test PATHTOSAVED_CONVBIG.pynet --width 0.1 --test-size 500 --test-batch-size 500

Note that "--update-test-net-name" will create a new model based on convBig and try to use the weights in the pickled PATHTOSAVED_CONVBIG.pynet to initialize that models weights. This is not always necessary, but is useful when the code for a model changes (in components) but does not effect the number or usage of weight, or when loading a model pickled by a cuda process into a cpu process.

The default specification type is the L_infinity Ball specified explicitly by "--spec boxSpec", which uses an epsilon specified by "--width"

The default specification type is the L_infinity Ball specified explicitly by "--spec boxSpec", which uses an epsilon specified by "--width"

Abstract Networks

Example Abstract Net

A cruical point of DiffAI v3 is that how a network is trained and abstracted should be part of the network description itself. In this release, we provide layers that allow one to alter how the abstraction works, in addition to providing a script for converting an abstract network to onyx so that the abstract analysis might be run in tensorflow. Below is a list of the abstract layers that we have included.

  • CorrMaxPool3D
  • CorrMaxPool2D
  • CorrFix
  • CorrMaxK
  • CorrRand
  • DecorrRand
  • DecorrMin
  • DeepLoss
  • ToZono
  • ToHZono
  • Concretize
  • CorrelateAll

Training Domain DSL

In DiffAI v3, a dsl has been provided to specify arbitrary training domains. In particular, it is now possible to train on combinations of attacks and abstract domains on specifications defined by attacks. Specifying training domains is possible in the command line using -d "DOMAIN_INITIALIZATION". The possible combinations are the classes listed in domains.py. The same syntax is also supported for testing domains, to allow for testing robustness with different epsilon-sized attacks and specifications.

Listed below are a few examples:

  • -t "IFGSM(k=4, w=0.1)" -t "ZNIPS(w=0.3)" Will first test with the PGD attack with an epsilon=w=0.1 and, the number of iterations k=4 and step size set to w/k. It will also test with the zonotope domain using the transformer specified in our NIPS 2018 paper with an epsilon=w=0.3.

  • -t "PGD(r=3,k=16,restart=2, w=0.1)" tests on points found using PGD with a step size of r*w/k and two restarts, and an attack-generated specification.

  • -d Point() is standard non-defensive training.

  • -d "LinMix(a=IFGSM(), b=Box(), aw=1, bw=0.1)" trains on points produced by pgd with the default parameters listed in domains.py, and points produced using the box domain. The loss is combined linearly using the weights aw and bw and scaled by 1/(aw + bw). The epsilon used for both is the ambient epsilon specified with "--width".

  • -d "DList((IFGSM(w=0.1),1), (Box(w=0.01),0.1), (Box(w=0.1),0.01))" is a generalization of the Mix domain allowing for training with arbitrarily many domains at once weighted by the given values (the resulting loss is scaled by the inverse of the sum of weights).

  • -d "AdvDom(a=IFGSM(), b=Box())" trains using the Box domain, but constructs specifications as Lāˆž balls containing the PGD attack image and the original image "o".

  • -d "BiAdv(a=IFGSM(), b=Box())" is similar, but creates specifications between the pgd attack image "a" and "o - (a - o)".

One domain we have found particularly useful for training is Mix(a=PGD(r=3,k=16,restart=2, w=0.1), b=BiAdv(a=IFGSM(k=5, w=0.05)), bw=0.1).

While the above domains are all deterministic (up to gpu error and shuffling orders), we have also implemented nondeterministic training domains:

  • -d "Coin(a=IFGSM(), b=Box(), aw=1, bw=0.1)" is like Mix, but chooses which domain to train a batch with by the probabilities determined by aw / (aw + bw) and bw / (aw + bw).

  • -d "DProb((IFGSM(w=0.1),1), (Box(w=0.01),0.1), (Box(w=0.1),0.01))" is to Coin what DList is to Mix.

  • -d AdvDom(a=IFGSM(), b=DList((PointB(),1), (PointA(), 1), (Box(), 0.2))) can be used to share attack images between multiple training types. Here an attack image "m" is found using PGD, then both the original image "o" and the attack image "m" are passed to DList which trains using three different ways: PointA trains with "o", PointB trains with "m", and Box trains on the box produced between them. This can also be used with Mix.

  • -d Normal(w=0.3) trains using images sampled from a normal distribution around the provided image using standard deviation w.

  • -d NormalAdv(a=IFGSM(), w=0.3) trains using PGD (but this could be an abstract domain) where perturbations are constrained to a box determined by a normal distribution around the original image with standard deviation w.

  • -d InSamp(0.2, w=0.1) uses Inclusion sampling as defined in the ArXiv paper.

There are more domains implemented than listed here, and of course more interesting combinations are possible. Please look carefully at domains.py for default values and further options.

Parameter Scheduling DSL

In place of many constants, you can use the following scheduling devices.

  • Lin(s,e,t,i) Linearly interpolates between s and e over t epochs, using s for the first i epochs.

  • Until(t,a,b) Uses a for the first t epochs, then switches to using b (telling b the current epoch starting from 0 at epoch t).

Suggested Training

LinMix(a=IFGSM(k=2), b=InSamp(Lin(0,1,150,10)), bw = Lin(0,0.5,150,10)) is a training goal that appears to work particularly well for CIFAR10 networks.

Contents

  • components.py: A high level neural network library for composable layers and operations
  • goals.py: The DSL for specifying training losses and domains, and attacks which can be used as a drop in replacement for pytorch tensors in any model built with components from components.py
  • scheduling.py: The DSL for specifying parameter scheduling.
  • models.py: A repository of models to train with which are used in the paper.
  • convert.py: A utility for converting a model with a training or testing domain (goal) into an onyx network. This is useful for exporting DiffAI abstractions to tensorflow.
  • __main__.py: The entry point to run the experiments.
  • helpers.py: Assorted helper functions. Does some monkeypatching, so you might want to be careful importing our library into your project.
  • AllExperimentsSerial.sh: A script which runs the training experiments from the 2019 ArXiv paper from table 4 and 5 and figure 5.

Notes

Not all of the datasets listed in the help message are supported. Supported datasets are:

  • CIFAR10
  • CIFAR100
  • MNIST
  • SVHN
  • FashionMNIST

Unsupported datasets will not necessarily throw errors.

Results on Standard Networks

Download all defended networks, logs, and configs

MNIST

Network Number of Neurons Number of Parameters Number ReLU Layers
FFNN 500 119910 5
ConvSmall 3604 89606 3
ConvMed 4804 166406 3
ConvBig 48064 1974762 6
ConvLargeIBP 175816 5426402 6
TruncatedVGG 151040 13109706 5

0.1

Config Log

Network Standard Accuracy MI_FGSM Accuracy HBox Provability
FFNN 93.3% 90.8% 88.9%
ConvSmall 97.8% 96.2% 95.5%
ConvMed 97.8% 96.3% 95.5%
ConvBig 98.5% 97.2% 95.6%
ConvLargeIBP 98.7% 97.5% 95.8%
TruncatedVGG 98.9% 97.7% 95.6%

0.3

Config Log

Network Standard Accuracy MI_FGSM Accuracy HBox Provability
FFNN 80.2% 73.4% 62.6%
ConvSmall 96.9% 93.6% 89.1%
ConvMed 96.6% 93.1% 89.3%
ConvBig 97.0% 95.2% 87.8%
ConvLargeIBP 97.2% 95.4% 88.8%
TruncatedVGG 96.5% 94.4% 87.6%

CIFAR10

Network Number of Neurons Number of Parameters Number ReLU Layers
FFNN 500 348710 5
ConvSmall 4852 125318 3
ConvMed 6244 214918 3
ConvBig 62464 2466858 6
ConvLargeIBP 229576 6963554 6
TruncatedVGG 197120 17043018 5

2/255

Config Log

Network Standard Accuracy MI_FGSM Accuracy HBox Provability
FFNN 45.1% 37.0% 33.1%
ConvSmall 56.1% 46.2% 42.4%
ConvMed 56.9% 46.6% 43.2%
ConvBig 61.9% 51.4% 45.0%
ConvLargeIBP 61.1% 51.4% 44.5%
TruncatedVGG 62.3% 51.4% 45.5%

8/255

Config Log

Network Standard Accuracy MI_FGSM Accuracy HBox Provability
FFNN 33.5% 23.8% 19.0%
ConvSmall 42.6% 30.5% 24.9%
ConvMed 43.6% 30.3% 24.7%
ConvBig 46.0% 34.2% 25.2%
ConvLargeIBP 46.2% 34.7% 27.2%
TruncatedVGG 45.9% 34.4% 27.0%

Reproducing Results

Download Defended Networks

All training runs from the paper can be reproduced as by the following command, in the same order as Table 6 in the appendix.

./AllExperimentsSerial.sh "-t MI_FGSM(k=20,r=2) -t HBox --test-size 10000 --test-batch-size 200 --test-freq 400 --save-freq 1 --epochs 420 --out all_experiments --write-first True --test-first False"

The training schemes can be written as follows (the names differ slightly from the presentation in the paper):

  • Baseline: LinMix(a=Point(), b=Box(w=Lin(0,0.031373,150,10)), bw=Lin(0,0.5,150,10))
  • InSamp: LinMix(a=Point(), b=InSamp(Lin(0,1,150,10)), bw=Lin(0,0.5, 150,10))
  • InSampLPA: LinMix(a=Point(), b=InSamp(Lin(0,1,150,20), w=Lin(0,0.031373, 150, 20)), bw=Lin(0,0.5, 150, 20))
  • Adv_{1}ISLPA: LinMix(a=IFGSM(w=Lin(0,0.031373,20,20), k=1), b=InSamp(Lin(0,1,150,10), w=Lin(0,0.031373,150,10)), bw=Lin(0,0.5,150,10))
  • Adv_{3}ISLPA: LinMix(a=IFGSM(w=Lin(0,0.031373,20,20), k=3), b=InSamp(Lin(0,1,150,10), w=Lin(0,0.031373,150,10)), bw=Lin(0,0.5,150,10))
  • Baseline_{18}: LinMix(a=Point(), b=InSamp(Lin(0,1,200,40)), bw=Lin(0,0.5,200,40))
  • InSamp_{18}: LinMix(a=IFGSM(w=Lin(0,0.031373,20,20)), b=InSamp(Lin(0,1,200,40)), bw=Lin(0,0.5,200,40))
  • Adv_{5}IS_{18}: LinMix(b=InSamp(Lin(0,1,200,40)), bw=Lin(0,0.5, 200, 40))
  • BiAdv_L: LinMix(a=IFGSM(k=2), b=BiAdv(a=IFGSM(k=3, w=Lin(0,0.031373, 150, 30)), b=Box()), bw=Lin(0,0.6, 200, 30))

To test a saved network as in the paper, use the following command:

python . -D CIFAR10 -n ResNetLarge_LargeCombo -d Point --width 0.031373 --normalize-layer True --clip-norm False -t 'MI_FGSM(k=20,r=2)' -t HBox --test-size 10000 --test-batch-size 200 --epochs 1 --test NAMEOFSAVEDNET.pynet 

Assorted

Citing This Framework

@inproceedings{
  title={Differentiable Abstract Interpretation for Provably Robust Neural Networks},
  author={Mirman, Matthew and Gehr, Timon and Vechev, Martin},
  booktitle={International Conference on Machine Learning (ICML)},
  year={2018},
  url={https://www.icml.cc/Conferences/2018/Schedule?showEvent=2477},
}

Contributors

License and Copyright

Note that the project description data, including the texts, logos, images, and/or trademarks, for each open source project belongs to its rightful owner. If you wish to add or remove any projects, please contact us at [email protected].