【Awesome】最全的机器学习可解释性资料(machine-learning-interpretability)

3 月 1 日 专知

【导读】 大家知道机器学习和深度学习在近几年已经取得了极大的发展,应用范围十分广泛,对各行各业也都产生了深远的影响。但是现在从事机器学习和深度学习的研究人员研究的焦点逐渐变成可解释性。什么是可解释性?可解释性就是不仅仅要知道怎么做,同时也要知道为什么这么做会产生这样的结果,不论结果是否理想。今天小编就给大家分享一些机器学习可解释性资料


下面是一个局部的机器学习蓝图,可以从总体上帮助降低做机器学习任务的风险程度。



▌资料目录




  • 全面的软件示例和教程(Comprehensive Software Examples and Tutorials)

  • 可解释性或合适的增强软件包Explainability- or Fairness-Enhancing Software Package

    • Browser

    • Python

    • R

  • 免费的书(Free Books)

  • 其他可解释性和合适的资源和列表

  • 论文(Review and General Papers)

  • 教学资源(Teaching Resources)

  • 可解释(“白盒”)或合适的建模包(Interpretable ("Whitebox") or Fair Modeling Packages)

    • C/C++

    • Python

    • R

Comprehensive Software Examples and Tutorials




  • Getting a Window into your Black Box Model

  • IML

  • Interpretable Machine Learning with Python

  • Interpreting Machine Learning Models with the iml Package

  • Machine Learning Explainability by Kaggle Learn

  • Model Interpretability with DALEX

  • Model Interpretation series by Dipanjan (DJ) Sarkar:

    • The Importance of Human Interpretable Machine Learning

    • Model Interpretation Strategies

    • Hands-on Machine Learning Model Interpretation

  • Partial Dependence Plots in R

  • Visualizing ML Models with LIME


Expalinability-or Fairness-Enhancing Software Packages




  • Browser

    • What-if Tool

  • Python

    • ‍‍aequitas

    • AI Fairness 360

    • anchor

    • casme

    • cleverhans

    • ContrastiveExplanation (Foil Trees)

    • deeplift

    • deepvis

    • eli5

    • fairml

    • fairness

    • Integrated-Gradients

    • lofo-importance

    • L2X

    • lime

    • PDPbox

    • pyBreakDown

    • PyCEbox

    • shap

    • Skater

    • rationale

    • tensorflow/lucid

    • tensorflow/model-analysis

    • Themis

    • themis-ml

    • treeinterpreter

    • woe

    • xai

  • R

    • ALEPlot

    • breakDown

    • DALEX

    • ExplainPrediction

    • featureImportance

    • forestmodel

    • fscaret

    • ICEbox

    • iml

    • lightgbmExplainer

    • lime

    • live

    • mcr

    • pdp

    • shapleyR

    • smbinning

    • vip

    • xgboostExplainer


▌Free Books




  • Beyond Explainability: A Practical Guide to Managing Risk in Machine Learning Models

  • Fairness and Machine Learning

  • Interpretable Machine Learning


▌Other Interpretability and Fairness Resources and Lists



  • 8 Principles of Responsible ML

  • An Introduction to Machine Learning Interpretability

  • Awesome interpretable machine learning ;)

  • Awesome machine learning operations

  • algoaware

  • criticalML

  • Fairness, Accountability, and Transparency in Machine Learning (FAT/ML) Scholarship

  • Machine Learning Ethics References

  • Machine Learning Interpretability Resources

  • MIT AI Ethics Reading Group

  • XAI Resources


▌Review and General Papers




  • A Comparative Study of Fairness-Enhancing Interventions in Machine Learning

  • A Survey Of Methods For Explaining Black Box Models

  • Challenges for Transparency

  • Explaining Explanations: An Approach to Evaluating Interpretability of Machine Learning

  • On the Art and Science of Machine Learning Explanations

  • On the Responsibility of Technologists: A Prologue and Primer

  • Please Stop Explaining Black Box Models for High-Stakes Decisions

  • The Mythos of Model Interpretability

  • The Promise and Peril of Human Evaluation for Model Interpretability

  • Towards A Rigorous Science of Interpretable Machine Learning

  • Trends and Trajectories for Explainable, Accountable and Intelligible Systems: An HCI Research Agenda


▌Review and General Papers




  • An Introduction to Data Ethics

  • Fairness in Machine Learning

  • Human-Center Machine Learning

  • Practical Model Interpretability


▌Interpretable("Whitebox") or Fair Modeing Packages




  • C/C++

    • Certifiably Optimal RulE ListS

  • Python

    • Bayesian Case Model

    • Bayesian Ors-Of-Ands

    • Bayesian Rule List (BRL)

    • fair-classification

    • Falling Rule List (FRL)

    • H2O-3

      • Penalized Generalized Linear Models

      • Sparse Principal Components (GLRM)

    • Monotonic XGBoost

    • pyGAM

    • Risk-SLIM

    • Scikit-learn

      • Decision Trees

      • Generalized Linear Models

      • Sparse Principal Components

    • sklearn-expertsys

    • skope-rules

    • Super-sparse Linear Integer models (SLIMs)

  • R

    • arules

    • Causal SVM

    • elasticnet

    • gam

    • glmnet

    • H2O-3

      • Penalized Generalized Linear Models

      • Sparse Principal Components (GLRM)

    • Monotonic XGBoost

    • quantreg

    • rpart

    • RuleFit

    • Scalable Bayesian Rule Lists (SBRL)



参考链接:https://github.com/jphall663/awesome-machine-learning-interpretability#comprehensive-software-examples-and-tutorials

https://github.com/h2oai/mli-resources



END-

专 · 知

专知《深度学习:算法到实战》课程全部完成!490+位同学在学习,现在报名,限时优惠!网易云课堂人工智能畅销榜首位!

欢迎微信扫一扫加入专知人工智能知识星球群,获取最新AI专业干货知识教程视频资料和与专家交流咨询!

请加专知小助手微信(扫一扫如下二维码添加),加入专知人工智能主题群,咨询《深度学习:算法到实战》课程,咨询技术商务合作~

请PC登录www.zhuanzhi.ai或者点击阅读原文,注册登录专知,获取更多AI知识资料!

点击“阅读原文”,了解报名专知《深度学习:算法到实战》课程

登录查看更多
点赞 0

Factorization Machines (FM) are only used in a narrow range of applications and are not part of the standard toolbox of machine learning models. This is a pity, because even though FMs are recognized as being very successful for recommender system type applications they are a general model to deal with sparse and high dimensional features. Our Factorization Machine implementation provides easy access to many solvers and supports regression, classification and ranking tasks. Such an implementation simplifies the use of FM's for a wide field of applications. This implementation has the potential to improve our understanding of the FM model and drive new development.

点赞 0
阅读1+
Top