Adversarial Machine Learning

• Adversarial Machine Learning is a branch of machine learning where the adversary tries to provide malicious input to the machine learning model with an intent to fool the model.
• In this project, an agent is acting as an adversary to a malware detection model trained using features extracted from benign and malware android apps.
• The agent is created using reinforcement learning where it is allowed to interact with the malware detection model with the intent of developing an optimal policy to fool the detection model.

Prerequisites

• numpy
• pandas
• seaborn
• matplotlib
• sklearn
• xlwt
• efficient-apriori

Installing

!pip install efficient-apriori

Deployment

No deployement needed.

• Colab Interactive jupyter notebook

Mount your Google Drive

from google.colab import drive
drive.mount('/content/drive/')

Check your Folder Data

!ls Drive/test

Upload code from your system

from google.colab import files
uploaded = files.upload()

Contributing

Details on the code of conduct, and the process for submitting pull requests to me will be updated soon.

Versioning

For the versions available, see the tags on this repository.

Authors

• Piyush Nikam