Accountable Systems Lab at CMU

aaai-2021-tutorial

Anupam Datta, Matt Fredrikson, Klas Leino, Kaiji Lu, Shayak Sen, Zifan Wang

lib-attribution

Library providing attribution methods for Keras models. Supports Tensorflow and Theano backends.

smt-layer

self-correcting-networks

tool-qii

Ongoing: QII Code originally from Datta-Sen-Zick Oakland'16, actively developed

artifact-proxyuse-ccs2017

Snapshot of the proxy-use-related implementations and experiments/results presented in the CCS 2017 paper.

tool-adfisher

Run information flow experiments on the Web

verified_rup

Checks DRUP & DRAT proofs against DIMACS source. Extracted from verified Why3 code.

apc-website

cmu-transparency.github.io

demo-scala-labels

Label tracking and label-based policy enforcement demo for the origin privacy project.

gloronet

Visit the demo here:

Keras-GAN

Keras implementations of Generative Adversarial Networks.

lib-iris

marketing

project-evaluate-attr

Evaluate attribution methods

project-tree-mi-attack

template-doc-latex

website-satc

data-consumer-expenditure

guide-coding-and-tools

Coding guides for various languages the group is using. Useful for smoother collaboration among team members and releasing code to the world.

lib-notebook-utils

Various small utilities for use in jupyter notebooks.