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 Been Kim

Acquisition of Chess Knowledge in AlphaZero


Nov 27, 2021
Thomas McGrath, Andrei Kapishnikov, Nenad Tomašev, Adam Pearce, Demis Hassabis, Been Kim, Ulrich Paquet, Vladimir Kramnik

* 69 pages, 44 figures 

  Access Paper or Ask Questions

Best of both worlds: local and global explanations with human-understandable concepts


Jun 16, 2021
Jessica Schrouff, Sebastien Baur, Shaobo Hou, Diana Mincu, Eric Loreaux, Ralph Blanes, James Wexler, Alan Karthikesalingam, Been Kim


  Access Paper or Ask Questions

DISSECT: Disentangled Simultaneous Explanations via Concept Traversals


May 31, 2021
Asma Ghandeharioun, Been Kim, Chun-Liang Li, Brendan Jou, Brian Eoff, Rosalind W. Picard


  Access Paper or Ask Questions

Debugging Tests for Model Explanations


Nov 10, 2020
Julius Adebayo, Michael Muelly, Ilaria Liccardi, Been Kim

* A shorter version of this work will appear at Neurips 2020 

  Access Paper or Ask Questions

Concept Bottleneck Models


Jul 09, 2020
Pang Wei Koh, Thao Nguyen, Yew Siang Tang, Stephen Mussmann, Emma Pierson, Been Kim, Percy Liang

* ICML 2020 

  Access Paper or Ask Questions

On Concept-Based Explanations in Deep Neural Networks


Oct 17, 2019
Chih-Kuan Yeh, Been Kim, Sercan O. Arik, Chun-Liang Li, Pradeep Ravikumar, Tomas Pfister


  Access Paper or Ask Questions

BIM: Towards Quantitative Evaluation of Interpretability Methods with Ground Truth


Jul 23, 2019
Mengjiao Yang, Been Kim


  Access Paper or Ask Questions

Towards Realistic Individual Recourse and Actionable Explanations in Black-Box Decision Making Systems


Jul 22, 2019
Shalmali Joshi, Oluwasanmi Koyejo, Warut Vijitbenjaronk, Been Kim, Joydeep Ghosh


  Access Paper or Ask Questions

Explaining Classifiers with Causal Concept Effect (CaCE)


Jul 16, 2019
Yash Goyal, Uri Shalit, Been Kim


  Access Paper or Ask Questions

Visualizing and Measuring the Geometry of BERT


Jun 06, 2019
Andy Coenen, Emily Reif, Ann Yuan, Been Kim, Adam Pearce, Fernanda Viégas, Martin Wattenberg

* 8 pages, 5 figures 

  Access Paper or Ask Questions

Do Neural Networks Show Gestalt Phenomena? An Exploration of the Law of Closure


Mar 21, 2019
Been Kim, Emily Reif, Martin Wattenberg, Samy Bengio


  Access Paper or Ask Questions

Automating Interpretability: Discovering and Testing Visual Concepts Learned by Neural Networks


Feb 07, 2019
Amirata Ghorbani, James Wexler, Been Kim


  Access Paper or Ask Questions

An Evaluation of the Human-Interpretability of Explanation


Jan 31, 2019
Isaac Lage, Emily Chen, Jeffrey He, Menaka Narayanan, Been Kim, Sam Gershman, Finale Doshi-Velez

* arXiv admin note: substantial text overlap with arXiv:1802.00682 

  Access Paper or Ask Questions

Human-in-the-Loop Interpretability Prior


Oct 30, 2018
Isaac Lage, Andrew Slavin Ross, Been Kim, Samuel J. Gershman, Finale Doshi-Velez

* To appear at NIPS 2018, selected for a spotlight. 13 pages (incl references and appendix) 

  Access Paper or Ask Questions

Sanity Checks for Saliency Maps


Oct 28, 2018
Julius Adebayo, Justin Gilmer, Michael Muelly, Ian Goodfellow, Moritz Hardt, Been Kim

* NIPS 2018 Camera Ready Version 

  Access Paper or Ask Questions

To Trust Or Not To Trust A Classifier


Oct 26, 2018
Heinrich Jiang, Been Kim, Melody Y. Guan, Maya Gupta

* NIPS 2018 

  Access Paper or Ask Questions

Interpreting Black Box Predictions using Fisher Kernels


Oct 23, 2018
Rajiv Khanna, Been Kim, Joydeep Ghosh, Oluwasanmi Koyejo


  Access Paper or Ask Questions

Local Explanation Methods for Deep Neural Networks Lack Sensitivity to Parameter Values


Oct 08, 2018
Julius Adebayo, Justin Gilmer, Ian Goodfellow, Been Kim

* Workshop Track International Conference on Learning Representations (ICLR) 

  Access Paper or Ask Questions

Proceedings of the 2018 ICML Workshop on Human Interpretability in Machine Learning (WHI 2018)


Jul 03, 2018
Been Kim, Kush R. Varshney, Adrian Weller


  Access Paper or Ask Questions

Evaluating Feature Importance Estimates


Jun 28, 2018
Sara Hooker, Dumitru Erhan, Pieter-Jan Kindermans, Been Kim

* presented at 2018 ICML Workshop on Human Interpretability in Machine Learning (WHI 2018), Stockholm, Sweden 

  Access Paper or Ask Questions

xGEMs: Generating Examplars to Explain Black-Box Models


Jun 22, 2018
Shalmali Joshi, Oluwasanmi Koyejo, Been Kim, Joydeep Ghosh


  Access Paper or Ask Questions

Interpretability Beyond Feature Attribution: Quantitative Testing with Concept Activation Vectors (TCAV)


Jun 07, 2018
Been Kim, Martin Wattenberg, Justin Gilmer, Carrie Cai, James Wexler, Fernanda Viegas, Rory Sayres


  Access Paper or Ask Questions

How do Humans Understand Explanations from Machine Learning Systems? An Evaluation of the Human-Interpretability of Explanation


Feb 02, 2018
Menaka Narayanan, Emily Chen, Jeffrey He, Been Kim, Sam Gershman, Finale Doshi-Velez


  Access Paper or Ask Questions

The (Un)reliability of saliency methods


Nov 02, 2017
Pieter-Jan Kindermans, Sara Hooker, Julius Adebayo, Maximilian Alber, Kristof T. Schütt, Sven Dähne, Dumitru Erhan, Been Kim


  Access Paper or Ask Questions

Learning how to explain neural networks: PatternNet and PatternAttribution


Oct 24, 2017
Pieter-Jan Kindermans, Kristof T. Schütt, Maximilian Alber, Klaus-Robert Müller, Dumitru Erhan, Been Kim, Sven Dähne


  Access Paper or Ask Questions

Proceedings of the 2017 ICML Workshop on Human Interpretability in Machine Learning (WHI 2017)


Aug 08, 2017
Been Kim, Dmitry M. Malioutov, Kush R. Varshney, Adrian Weller


  Access Paper or Ask Questions

SmoothGrad: removing noise by adding noise


Jun 12, 2017
Daniel Smilkov, Nikhil Thorat, Been Kim, Fernanda Viégas, Martin Wattenberg

* 10 pages 

  Access Paper or Ask Questions

Towards A Rigorous Science of Interpretable Machine Learning


Mar 02, 2017
Finale Doshi-Velez, Been Kim


  Access Paper or Ask Questions

Proceedings of NIPS 2016 Workshop on Interpretable Machine Learning for Complex Systems


Nov 28, 2016
Andrew Gordon Wilson, Been Kim, William Herlands

* 31 papers 

  Access Paper or Ask Questions