Displaying 1 to 12 from 12 results

lucid - A collection of infrastructure and tools for research in neural network interpretability.

  •    Jupyter

Lucid is a collection of infrastructure and tools for research in neural network interpretability. In particular, it provides state of the art implementations of feature visualization techniques, and flexible abstractions that make it very easy to explore new research directions.

interpret - Fit interpretable models. Explain blackbox machine learning.

  •    C++

Historically, the most intelligible models were not very accurate, and the most accurate models were not intelligible. Microsoft Research has developed an algorithm called the Explainable Boosting Machine (EBM)* which has both high accuracy and intelligibility. EBM uses modern machine learning techniques like bagging and boosting to breathe new life into traditional GAMs (Generalized Additive Models). This makes them as accurate as random forests and gradient boosted trees, and also enhances their intelligibility and editability. In addition to EBM, InterpretML also supports methods like LIME, SHAP, linear models, partial dependence, decision trees and rule lists. The package makes it easy to compare and contrast models to find the best one for your needs.

DALEX - Descriptive mAchine Learning EXplanations

  •    R

Machine Learning models are widely used and have various applications in classification or regression tasks. Due to increasing computational power, availability of new data sources and new methods, ML models are more and more complex. Models created with techniques like boosting, bagging of neural networks are true black boxes. It is hard to trace the link between input variables and model outcomes. They are use because of high performance, but lack of interpretability is one of their weakest sides. In many applications we need to know, understand or prove how input variables are used in the model and what impact do they have on final model prediction. DALEX is a set of tools that help to understand how complex models are working.




mli-resources - Machine Learning Interpretability Resources

  •    Jupyter

Machine learning algorithms create potentially more accurate models than linear models, but any increase in accuracy over more traditional, better-understood, and more easily explainable techniques is not practical for those who must explain their models to regulators or customers. For many decades, the models created by machine learning algorithms were generally taken to be black-boxes. However, a recent flurry of research has introduced credible techniques for interpreting complex, machine-learned models. Materials presented here illustrate applications or adaptations of these techniques for practicing data scientists. Want to contribute your own examples? Just make a pull request.

text_nn - Text classification models. Used a submodule for other projects.

  •    Python

This repository assumes glove embeddings. Download Glove embeddings at: https://nlp.stanford.edu/projects/glove/ And place glove.6B/glove.6B.300d.txt in data/embeddings/glove.6B/glove.6B.300d.txt. This code supports the the NewsGroup dataset and the BeerReview dataset. The for access to the BeerReview and the corresponding embeddings, please contact me or Tao Lei. I've included the NewsGroup dataset, conveiently provided by SKLearn so you can run code out of the box.

bayesgrad - BayesGrad: Explaining Predictions of Graph Convolutional Networks

  •    Jupyter

The paper is available on arXiv, https://arxiv.org/abs/1807.01985. From left: tox21 pyridine (C5H5N), tox21 SR-MMP, delaney solubility visualization.

breakDown - Model Agnostics breakDown plots

  •    R

The breakDown package is a model agnostic tool for decomposition of predictions from black boxes. Break Down Table shows contributions of every variable to a final prediction. Break Down Plot presents variable contributions in a concise graphical way. This package works for binary classifiers and general regression models.


diabetes_use_case - Sample use case for Xavier AI in Healthcare conference: https://www

  •    Jupyter

Recent advances enable practitioners to break open machine learning’s “black box”. From machine learning algorithms guiding analytical tests in drug manufacture, to predictive models recommending courses of treatment, to sophisticated software that can read images better than doctors, machine learning has promised a new world of healthcare where algorithms can assist, or even outperform, professionals in consistency and accuracy, saving money and avoiding potentially life-threatening mistakes. But what if your doctor told you that you were sick but could not tell you why? Imagine a hospital that hospitalized and discharged patients but was unable to provide specific justification for these decisions. For decades, this was a roadblock for the adoption of machine learning algorithms in healthcare: they could make data-driven decisions that helped practitioners, payers, and patients, but they couldn’t tell users why those decisions were made.

interpretable_machine_learning_with_python - Practical techniques for interpreting machine learning models

  •    Jupyter

Monotonicity constraints can turn opaque, complex models into transparent, and potentially regulator-approved models, by ensuring predictions only increase or only decrease for any change in a given input variable. In this notebook, I will demonstrate how to use monotonicity constraints in the popular open source gradient boosting package XGBoost to train a simple, accurate, nonlinear classifier on the UCI credit card default data. Once we have trained a monotonic XGBoost model, we will use partial dependence plots and individual conditional expectation (ICE) plots to investigate the internal mechanisms of the model and to verify its monotonic behavior. Partial dependence plots show us the way machine-learned response functions change based on the values of one or two input variables of interest, while averaging out the effects of all other input variables. ICE plots can be used to create more localized descriptions of model predictions, and ICE plots pair nicely with partial dependence plots. An example of generating regulator mandated reason codes from high fidelity Shapley explanations for any model prediction is also presented. The combination of monotonic XGBoost, partial dependence, ICE, and Shapley explanations is likely the most direct way to create an interpretable machine learning model today.

shapeshop - Towards Understanding Deep Learning Representations via Interactive Experimentation

  •    Python

The Shape Workshop (ShapeShop) is an interactive system for visualizing and understanding what representations a neural network model has learned in images consisting of simple shapes. It encourages model building, experimentation, and comparison to helps users explore the robustness of image classifiers. Read the paper. View the poster. Watch the teaser.