Merge pull request #192 from swag2198/master

Adding explanation certification (ICML 2024)

Author: vijay-arya

.github/workflows/Build.yml

@@ -389,6 +389,38 @@ jobs:
       - name: Step 6 - Test IMD Notebook
         run: pytest --nbmake ./examples/imd/imd_example.ipynb
 
+  build-ecertify-on-py38-310:
+    # The type of runner that the job will run on
+    runs-on: "${{ matrix.os }}"
+    strategy:
+      fail-fast: false
+      matrix:
+        #        os: [ubuntu-18.04, ubuntu-latest, macos-latest, windows-latest]
+        os: [ubuntu-20.04, macos-latest, windows-latest]
+        python-version: ["3.10"]
+
+    # Steps represent a sequence of tasks that will be executed as part of the job
+    steps:
+      - name: Step 1 - checkout aix360 repository
+        uses: actions/checkout@v3
+
+      - name: Step 2 - set up python version
+        uses: actions/setup-python@v4
+        with:
+          python-version: "${{ matrix.python-version }}"
+
+      - name: Step 3 - upgrade setuptools
+        run: pip3 install pytest nbmake wheel --upgrade setuptools
+
+      - name: Step 4 - Install aix360 with ecertify algorithm related dependencies
+        run: pip3 install .[ecertify]
+
+      - name: Step 5 - Test Ecertify
+        run: python ./tests/ecertify/test_ecertify.py
+
+      - name: Step 6 - Test Ecertify Notebook
+        run: pytest --nbmake ./examples/ecertify/certification_example_fico.ipynb
+
   # job to build groupedce algorithm on python 3.10.
   build-groupedce-on-py310:
     # The type of runner that the job will run on

README.md

@@ -40,9 +40,12 @@ We have developed the package with extensibility in mind. This library is still
 - Teaching AI to Explain its Decisions ([Hind et al., 2019](https://doi.org/10.1145/3306618.3314273)) 
 - Order Constraints in Optimal Transport ([Lim et al.,2022](https://arxiv.org/abs/2110.07275), [Github](https://github.com/IBM/otoc))
 
+### Certifying local explanations
+- Trust Regions for Explanations via Black-Box Probabilistic Certification (Ecertify) ([Dhurandhar et al., 2024](https://arxiv.org/abs/2402.11168))
+
 ### Global direct explanations
 
-- Interpretable Model Differencing (IMD) ([Haldar et al., 2023](https://arxiv.org/abs/2306.06473))
+- Interpretable Model Differencing (IMD) ([Haldar et al., 2023](https://proceedings.mlr.press/v216/haldar23a.html))
 - CoFrNets (Continued Fraction Nets) ([Puri et al., 2021](https://papers.nips.cc/paper/2021/file/b538f279cb2ca36268b23f557a831508-Paper.pdf))
 - Boolean Decision Rules via Column Generation (Light Edition) ([Dash et al., 2018](https://papers.nips.cc/paper/7716-boolean-decision-rules-via-column-generation))
 - Generalized Linear Rule Models ([Wei et al., 2019](http://proceedings.mlr.press/v97/wei19a.html))
@@ -67,6 +70,7 @@ We have developed the package with extensibility in mind. This library is still
 | contrastive    |cem, cem_maf    | macOS, Ubuntu, Windows        | 3.6  |
 | dipvae         | dipvae| macOS, Ubuntu, Windows        | 3.10 |
 | gce            | gce | macOS, Ubuntu, Windows        | 3.10 |
+| ecertify       | ecertify | macOS, Ubuntu, Windows        | 3.10 |
 | imd            | imd | macOS, Ubuntu                 | 3.10 |
 | lime           | lime| macOS, Ubuntu, Windows        | 3.10 |
 | matching       | matching| macOS, Ubuntu, Windows        | 3.10 |