Skip to content

A framework for generating probably approximately correct (PAC) and local explanations of black-box machine learning models based on SyGuS (Syntax-Guided Synthesis).

Notifications You must be signed in to change notification settings

bishwamittra/pacX

Repository files navigation

Introduction

A framework for generating probably approximately correct (PAC) and local explanations of black-box machine learning models based on SyGuS (Syntax-Guided Synthesis).

Install

  • run pip install -r requirements.txt to install required python libraries
  • CVC4

Documentation

Tests

All test files are under tests directory

  • python -m tests.test_sygus

Citations

@inproceedings{ghosh2020formal,
author={Neider, Daniel and Ghosh, Bishwamittra},
title={Probably Approximately Correct Explanations of Machine Learning Models via Syntax-Guided Synthesis},
booktitle={arXiv:2009.08770},
month={9},
year={2020}}

Contact

Bishwamittra Ghosh (bghosh@u.nus.edu)

Issues, questions, bugs, etc.

Please click on "issues" at the top and create a new issue. All issues are responded to promptly.

About

A framework for generating probably approximately correct (PAC) and local explanations of black-box machine learning models based on SyGuS (Syntax-Guided Synthesis).

Resources

Stars

Watchers

Forks

Packages

No packages published