Get our free extension to see links to code for papers anywhere online!

Chrome logo Add to Chrome

Firefox logo Add to Firefox
Picture for Sarah Tan

Sarah Tan

Distilling Heterogeneity: From Explanations of Heterogeneous Treatment Effect Models to Interpretable Policies


Nov 05, 2021
Han Wu, Sarah Tan, Weiwei Li, Mia Garrard, Adam Obeng, Drew Dimmery, Shaun Singh, Hanson Wang, Daniel Jiang, Eytan Bakshy

Figure 1 for Distilling Heterogeneity: From Explanations of Heterogeneous Treatment Effect Models to Interpretable Policies
Figure 2 for Distilling Heterogeneity: From Explanations of Heterogeneous Treatment Effect Models to Interpretable Policies
Figure 3 for Distilling Heterogeneity: From Explanations of Heterogeneous Treatment Effect Models to Interpretable Policies
Figure 4 for Distilling Heterogeneity: From Explanations of Heterogeneous Treatment Effect Models to Interpretable Policies
* A short version was presented at MIT CODE 2021 
  Access Paper or Ask Questions

How Interpretable and Trustworthy are GAMs?


Jun 11, 2020
Chun-Hao Chang, Sarah Tan, Ben Lengerich, Anna Goldenberg, Rich Caruana

Figure 1 for How Interpretable and Trustworthy are GAMs?
Figure 2 for How Interpretable and Trustworthy are GAMs?
Figure 3 for How Interpretable and Trustworthy are GAMs?
Figure 4 for How Interpretable and Trustworthy are GAMs?

  Access Paper or Ask Questions

Purifying Interaction Effects with the Functional ANOVA: An Efficient Algorithm for Recovering Identifiable Additive Models


Nov 12, 2019
Benjamin Lengerich, Sarah Tan, Chun-Hao Chang, Giles Hooker, Rich Caruana

Figure 1 for Purifying Interaction Effects with the Functional ANOVA: An Efficient Algorithm for Recovering Identifiable Additive Models
Figure 2 for Purifying Interaction Effects with the Functional ANOVA: An Efficient Algorithm for Recovering Identifiable Additive Models
Figure 3 for Purifying Interaction Effects with the Functional ANOVA: An Efficient Algorithm for Recovering Identifiable Additive Models
Figure 4 for Purifying Interaction Effects with the Functional ANOVA: An Efficient Algorithm for Recovering Identifiable Additive Models

  Access Paper or Ask Questions

"Why Should You Trust My Explanation?" Understanding Uncertainty in LIME Explanations


Jun 04, 2019
Yujia Zhang, Kuangyan Song, Yiming Sun, Sarah Tan, Madeleine Udell

Figure 1 for "Why Should You Trust My Explanation?" Understanding Uncertainty in LIME Explanations
Figure 2 for "Why Should You Trust My Explanation?" Understanding Uncertainty in LIME Explanations
Figure 3 for "Why Should You Trust My Explanation?" Understanding Uncertainty in LIME Explanations
Figure 4 for "Why Should You Trust My Explanation?" Understanding Uncertainty in LIME Explanations

  Access Paper or Ask Questions

Interpretability is Harder in the Multiclass Setting: Axiomatic Interpretability for Multiclass Additive Models


Oct 22, 2018
Xuezhou Zhang, Sarah Tan, Paul Koch, Yin Lou, Urszula Chajewska, Rich Caruana

Figure 1 for Interpretability is Harder in the Multiclass Setting: Axiomatic Interpretability for Multiclass Additive Models
Figure 2 for Interpretability is Harder in the Multiclass Setting: Axiomatic Interpretability for Multiclass Additive Models
Figure 3 for Interpretability is Harder in the Multiclass Setting: Axiomatic Interpretability for Multiclass Additive Models
Figure 4 for Interpretability is Harder in the Multiclass Setting: Axiomatic Interpretability for Multiclass Additive Models
* Preprint 
  Access Paper or Ask Questions

Distill-and-Compare: Auditing Black-Box Models Using Transparent Model Distillation


Oct 11, 2018
Sarah Tan, Rich Caruana, Giles Hooker, Yin Lou

Figure 1 for Distill-and-Compare: Auditing Black-Box Models Using Transparent Model Distillation
Figure 2 for Distill-and-Compare: Auditing Black-Box Models Using Transparent Model Distillation
Figure 3 for Distill-and-Compare: Auditing Black-Box Models Using Transparent Model Distillation
Figure 4 for Distill-and-Compare: Auditing Black-Box Models Using Transparent Model Distillation
* Camera-ready version for AAAI/ACM AIES 2018. Data and pseudocode at https://github.com/shftan/auditblackbox. Previously titled "Detecting Bias in Black-Box Models Using Transparent Model Distillation". A short version was presented at NIPS 2017 Symposium on Interpretable Machine Learning 
  Access Paper or Ask Questions

Investigating Human + Machine Complementarity for Recidivism Predictions


Aug 28, 2018
Sarah Tan, Julius Adebayo, Kori Inkpen, Ece Kamar

Figure 1 for Investigating Human + Machine Complementarity for Recidivism Predictions
Figure 2 for Investigating Human + Machine Complementarity for Recidivism Predictions
Figure 3 for Investigating Human + Machine Complementarity for Recidivism Predictions
Figure 4 for Investigating Human + Machine Complementarity for Recidivism Predictions

  Access Paper or Ask Questions

Transparent Model Distillation


Jan 26, 2018
Sarah Tan, Rich Caruana, Giles Hooker, Albert Gordo

Figure 1 for Transparent Model Distillation
Figure 2 for Transparent Model Distillation
Figure 3 for Transparent Model Distillation
Figure 4 for Transparent Model Distillation

  Access Paper or Ask Questions

A Double Parametric Bootstrap Test for Topic Models


Nov 21, 2017
Skyler Seto, Sarah Tan, Giles Hooker, Martin T. Wells

Figure 1 for A Double Parametric Bootstrap Test for Topic Models
Figure 2 for A Double Parametric Bootstrap Test for Topic Models
Figure 3 for A Double Parametric Bootstrap Test for Topic Models
Figure 4 for A Double Parametric Bootstrap Test for Topic Models
* Presented at NIPS 2017 Symposium on Interpretable Machine Learning 
  Access Paper or Ask Questions