2020

Explainable Active Learning (XAL): An Empirical Study of How Local Explanations Impact Annotator Experience

Ghai, Bhavya, Liao, Q. Vera, Zhang, Yunfeng et al.

Understand

The wide adoption of Machine Learning technologies has created a rapidly growing demand for people who can train ML models.

Some advocated the term "machine teacher" to refer to the role of people who inject domain knowledge into ML models.
One promising learning paradigm is Active Learning (AL), by which the model intelligently selects instances to query the machine teacher for labels.
However, in current AL settings, the human-AI interface remains minimal and opaque.

Built on

The need for cognition
John T Cacioppo and Richard E Petty. 1982 · 1982
Earlier work this paper cites.
Effects of need for cognition on message evaluation, recall, and persuasion
John T Cacioppo, Richard E Petty, and Katherine J Morris. 1983 · 1983
Earlier work this paper cites.
Personality and persuasion: Need for cognition moderates the persistence and resistance of attitude changes
Curtis P Haugtvedt and Richard E Petty. 1992 · 1992
Earlier work this paper cites.
Query by committee. In Proceedings of the fifth annual workshop on Computational learning theory . ACM, 287–294
H Sebastian Seung, Manfred Opper, and Haim Sompolinsky. 1992 · 1992
Earlier work this paper cites.
Improving generalization with active learning
David Cohn, Les Atlas, and Richard Ladner. 1994 · 1994
Earlier work this paper cites.
Computer usability satisfaction questionnaires: Psychometric evaluation and instructions for use
JR Lewis. 1995 · 1995
Earlier work this paper cites.
Adult Income dataset (UCI Machine Learning Repository)
Ronny Kohavi and Barry Becker. 1996 · 1996
Earlier work this paper cites.
Selective sampling using the query by committee algorithm
Yoav Freund, H Sebastian Seung, Eli Shamir, and Naftali Tishby. 1997 · 1997
Earlier work this paper cites.
Consensually driven explanation in science teaching
Karen Meyer and Earl Woodruff. 1997 · 1997
Earlier work this paper cites.
Initial trust formation in new organizational relationships
D Harrison McKnight, Larry L Cummings, and Norman L Chervany. 1998 · 1998
Earlier work this paper cites.
Utopian: User-driven topic modeling based on interactive nonnegative matrix factorization
Jaegul Choo, Changhyun Lee, Chandan K Reddy, and Haesun Park. 2013 · 2001
Earlier work this paper cites.
Developing and validating trust measures for e-commerce: An integrative typology
D Harrison McKnight, Vivek Choudhury, and Charles Kacmar. 2002 · 2002
Earlier work this paper cites.
On active learning for data acquisition. In 2002 IEEE International Conference on Data Mining, 2002. Proceedings. IEEE, 562–569
Zhiqiang Zheng and Balaji Padmanabhan. 2002 · 2002
Earlier work this paper cites.
Optimal Motion Control of a Ground Vehicle
David A. Anisi. 2003 · 2003
Earlier work this paper cites.
Interactive machine learning. In Proceedings of the 8th international conference on Intelligent user interfaces . ACM, 39–45
Jerry Alan Fails and Dan R Olsen Jr. 2003 · 2003
Earlier work this paper cites.
The Sage dictionary of statistics: a practical resource for students in the social sciences
Duncan Cramer and Dennis Laurence Howitt. 2004 · 2004
Earlier work this paper cites.
Theory of mind for learning and teaching: The nature and role of explanation
Henry M Wellman and Kristin H Lagattuta. 2004 · 2004
Earlier work this paper cites.
Reducing labeling effort for structured prediction tasks. In AAAI , Vol. 5. 746–751
Aron Culotta and Andrew McCallum. 2005 · 2005
Earlier work this paper cites.
Semi-supervised learning literature survey
Xiaojin Jerry Zhu. 2005 · 2005
Earlier work this paper cites.
Active learning with feedback on features and instances
Hema Raghavan, Omid Madani, and Rosie Jones. 2006 · 2006
Earlier work this paper cites.
Open user profiles for adaptive news systems: help or harm?. In Proceedings of the 16th international conference on World Wide Web . 11–20
Jae-wook Ahn, Peter Brusilovsky, Jonathan Grady, Daqing He, and Sue Yeon Syn. 2007 · 2007
Earlier work this paper cites.
Margin based active learning. In International Conference on Computational Learning Theory . Springer, 35–50
Maria-Florina Balcan, Andrei Broder, and Tong Zhang. 2007 · 2007
Earlier work this paper cites.
Active class selection. In European Conference on Machine Learning . Springer, 640–647
Rachel Lomasky, Carla E Brodley, Matthew Aernecke, David Walt, and Mark Friedl. 2007 · 2007
Earlier work this paper cites.
Toward harnessing user feedback for machine learning. In Proceedings of the 12th international conference on Intelligent user interfaces . ACM, 82–91
Simone Stumpf, Vidya Rajaram, Lida Li, Margaret Burnett, Thomas Dietterich, Erin Sullivan, Russell Drummond, and Jonathan Herlocker. 2007 · 2007
Earlier work this paper cites.
Hierarchical sampling for active learning. In Proceedings of the 25th international conference on Machine learning . ACM, 208–215
Sanjoy Dasgupta and Daniel Hsu. 2008 · 2008
Earlier work this paper cites.
Proactive learning: cost-sensitive active learning with multiple imperfect oracles. In Proceedings of the 17th ACM conference on Information and knowledge management . 619–628
Pinar Donmez and Jaime G Carbonell. 2008 · 2008
Earlier work this paper cites.
CueFlik: interactive concept learning in image search. In Proceedings of the sigchi conference on human factors in computing systems . 29–38
James Fogarty, Desney Tan, Ashish Kapoor, and Simon Winder. 2008 · 2008
Earlier work this paper cites.
An analysis of active learning strategies for sequence labeling tasks. In Proceedings of the conference on empirical methods in natural language processing . Association for Computational Linguistics, 1070–1079
Burr Settles and Mark Craven. 2008 · 2008
Earlier work this paper cites.
Active learning by labeling features. In Proceedings of the 2009 Conference on Empirical Methods in Natural Language Processing: Volume 1-Volume 1 . Association for Computational Linguistics, 81–90
Gregory Druck, Burr Settles, and Andrew McCallum. 2009 · 2009
Earlier work this paper cites.
Why and why not explanations improve the intelligibility of context-aware intelligent systems. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems . 2119–2128
Brian Y Lim, Anind K Dey, and Daniel Avrahami. 2009 · 2009
Earlier work this paper cites.
Active learning literature survey
Burr Settles. 2009 · 2009
Earlier work this paper cites.
Interacting meaningfully with machine learning systems: Three experiments
Simone Stumpf, Vidya Rajaram, Lida Li, Weng-Keen Wong, Margaret Burnett, Thomas Dietterich, Erin Sullivan, and Jonathan Herlocker. 2009 · 2009
Earlier work this paper cites.
EnsembleMatrix: interactive visualization to support machine learning with multiple classifiers. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems . 1283–1292
Justin Talbot, Bongshin Lee, Ashish Kapoor, and Desney S Tan. 2009 · 2009
Earlier work this paper cites.

Similar

Designing interactions for robot active learners
Maya Cakmak, Crystal Chao, and Andrea L Thomaz. 2010 · 2010
Cited alongside, same era.
Transparent active learning for robots. In 2010 5th ACM/IEEE International Conference on Human-Robot Interaction (HRI) . IEEE, 317–324
Crystal Chao, Maya Cakmak, and Andrea L Thomaz. 2010 · 2010
Cited alongside, same era.
Active learning by querying informative and representative examples. In Advances in neural information processing systems . 892–900
Sheng-Jun Huang, Rong Jin, and Zhi-Hua Zhou. 2010 · 2010
Cited alongside, same era.
Interactive optimization for steering machine classification. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems . 1343–1352
Ashish Kapoor, Bongshin Lee, Desney Tan, and Eric Horvitz. 2010 · 2010
Cited alongside, same era.
Toolkit to support intelligibility in context-aware applications. In Proceedings of the 12th ACM international conference on Ubiquitous computing . 13–22
Brian Y Lim and Anind K Dey. 2010 · 2010
Cited alongside, same era.
Towards maximizing the accuracy of human-labeled sensor data. In Proceedings of the 15th international conference on Intelligent user interfaces . ACM, 259–268
Stephanie L Rosenthal and Anind K Dey. 2010 · 2010
Cited alongside, same era.
Prototype selection for interpretable classification
Jacob Bien, Robert Tibshirani, et al · 2011
Cited alongside, same era.
Why-oriented end-user debugging of naive Bayes text classification
Todd Kulesza, Simone Stumpf, Weng-Keen Wong, Margaret M Burnett, Stephen Perona, Andrew Ko, and Ian Oberst. 2011 · 2011
Cited alongside, same era.
Generation of meaningful robot expressions with active learning. In 2011 6th ACM/IEEE International Conference on Human-Robot Interaction (HRI) . IEEE, 243–244
Giovanni Saponaro and Alexandre Bernardino. 2011 · 2011
Cited alongside, same era.
Closing the loop: Fast, interactive semi-supervised annotation with queries on features and instances. In Proceedings of the conference on empirical methods in natural language processing . Association for Computational Linguistics, 1467–1478
Burr Settles. 2011 · 2011
Cited alongside, same era.
Designing robot learners that ask good questions. In Proceedings of the seventh annual ACM/IEEE international conference on Human-Robot Interaction . ACM, 17–24
Maya Cakmak and Andrea L Thomaz. 2012 · 2012
Cited alongside, same era.
Explanation and Abductive Inference
Tania Lombrozo. 2012 · 2012
Cited alongside, same era.

Then

Theoretical considerations and development of a questionnaire to measure trust in automation. In Congress of the International Ergonomics Association . Springer, 13–30
Moritz Körber. 2018 · 2018
Later among the works it cites.
The mythos of model interpretability
Zachary C Lipton. 2018 · 2018
Later among the works it cites.
A Multidisciplinary Survey and Framework for Design and Evaluation of Explainable AI Systems
Sina Mohseni, Niloofar Zarei, and Eric D Ragan. 2018 · 2018
Later among the works it cites.
How do humans understand explanations from machine learning systems? an evaluation of the human-interpretability of explanation
Original
Menaka Narayanan, Emily Chen, Jeffrey He, Been Kim, Sam Gershman, and Finale Doshi-Velez. 2018 · 2018
Later among the works it cites.
A practical approach to measuring user engagement with the refined user engagement scale (UES) and new UES short form
Heather L O’Brien, Paul Cairns, and Mark Hall. 2018 · 2018
Later among the works it cites.
Manipulating and measuring model interpretability
Original
Forough Poursabzi-Sangdeh, Daniel G Goldstein, Jake M Hofman, Jennifer Wortman Vaughan, and Hanna Wallach. 2018 · 2018
Later among the works it cites.
Anchors: High-precision model-agnostic explanations. In Thirty-Second AAAI Conference on Artificial Intelligence
Marco Tulio Ribeiro, Sameer Singh, and Carlos Guestrin. 2018 · 2018
Later among the works it cites.
Closing the loop: User-centered design and evaluation of a human-in-the-loop topic modeling system. In 23rd International Conference on Intelligent User Interfaces . 293–304
Alison Smith, Varun Kumar, Jordan Boyd-Graber, Kevin Seppi, and Leah Findlater. 2018 · 2018
Later among the works it cites.
Active learning with confidence-based answers for crowdsourcing labeling tasks
Jinhua Song, Hao Wang, Yang Gao, and Bo An. 2018 · 2018
Later among the works it cites.
" Why Should I Trust Interactive Learners?" Explaining Interactive Queries of Classifiers to Users
Original
Stefano Teso and Kristian Kersting. 2018 · 2018
Later among the works it cites.
Interpretable to whom? A role-based model for analyzing interpretable machine learning systems
Original
Richard Tomsett, Dave Braines, Dan Harborne, Alun Preece, and Supriyo Chakraborty. 2018 · 2018
Later among the works it cites.
A benchmark and comparison of active learning for logistic regression
Yazhou Yang and Marco Loog. 2018 · 2018
Later among the works it cites.
Manifold: A model-agnostic framework for interpretation and diagnosis of machine learning models
Jiawei Zhang, Yang Wang, Piero Molino, Lezhi Li, and David S Ebert. 2018 · 2018
Later among the works it cites.
US Inflation Calculator
2019 (accessed July, 2019) · 2019
Later among the works it cites.
One Explanation Does Not Fit All: A Toolkit and Taxonomy of AI Explainability Techniques
Original
Vijay Arya, Rachel KE Bellamy, Pin-Yu Chen, Amit Dhurandhar, Michael Hind, Samuel C Hoffman, Stephanie Houde, Q Vera Liao, Ronny Luss, Aleksandra Mojsilović, et al · 2019
Later among the works it cites.
Machine Learning Interpretability: A Survey on Methods and Metrics
Diogo V Carvalho, Eduardo M Pereira, and Jaime S Cardoso. 2019 · 2019
Later among the works it cites.
Explaining Decision-Making Algorithms through UI: Strategies to Help Non-Expert Stakeholders. In Proceedings of the 2019 CHI Conference on Human Factors in Computing Systems . ACM, 559
Hao-Fei Cheng, Ruotong Wang, Zheng Zhang, Fiona O’Connell, Terrance Gray, F Maxwell Harper, and Haiyi Zhu. 2019 · 2019
Later among the works it cites.
Explaining models: an empirical study of how explanations impact fairness judgment. In Proceedings of the 24th International Conference on Intelligent User Interfaces . ACM, 275–285
Jonathan Dodge, Q Vera Liao, Yunfeng Zhang, Rachel KE Bellamy, and Casey Dugan. 2019 · 2019
Later among the works it cites.
A survey of methods for explaining black box models
Riccardo Guidotti, Anna Monreale, Salvatore Ruggieri, Franco Turini, Fosca Giannotti, and Dino Pedreschi. 2019 · 2019
Later among the works it cites.
Gamut: A design probe to understand how data scientists understand machine learning models. In Proceedings of the 2019 CHI Conference on Human Factors in Computing Systems . ACM, 579
Fred Hohman, Andrew Head, Rich Caruana, Robert DeLine, and Steven M Drucker. 2019 · 2019
Later among the works it cites.
Will you accept an imperfect ai? exploring designs for adjusting end-user expectations of ai systems. In Proceedings of the 2019 CHI Conference on Human Factors in Computing Systems . 1–14
Rafal Kocielnik, Saleema Amershi, and Paul N Bennett. 2019 · 2019
Later among the works it cites.
On Human Predictions with Explanations and Predictions of Machine Learning Models: A Case Study on Deception Detection. In Proceedings of the Conference on Fairness, Accountability, and Transparency . ACM, 29–38
Vivian Lai and Chenhao Tan. 2019 · 2019
Later among the works it cites.
Why these Explanations? Selecting Intelligibility Types for Explanation Goals.. In IUI Workshops
Brian Y Lim, Qian Yang, Ashraf M Abdul, and Danding Wang. 2019 · 2019
Later among the works it cites.
Explainability in human–agent systems
Avi Rosenfeld and Ariella Richardson. 2019 · 2019
Later among the works it cites.
Progressive disclosure: empirically motivated approaches to designing effective transparency. In Proceedings of the 24th International Conference on Intelligent User Interfaces . ACM, 107–120
Aaron Springer and Steve Whittaker. 2019 · 2019
Later among the works it cites.
Designing theory-driven user-centric explainable AI. In Proceedings of the 2019 CHI conference on human factors in computing systems . 1–15
Danding Wang, Qian Yang, Ashraf Abdul, and Brian Y Lim. 2019 · 2019
Later among the works it cites.
The What-If Tool: Interactive probing of machine learning models
James Wexler, Mahima Pushkarna, Tolga Bolukbasi, Martin Wattenberg, Fernanda Viégas, and Jimbo Wilson. 2019 · 2019
Later among the works it cites.
Local Decision Pitfalls in Interactive Machine Learning: An Investigation into Feature Selection in Sentiment Analysis
Tongshuang Wu, Daniel S Weld, and Jeffrey Heer. 2019 · 2019
Later among the works it cites.
Proxy tasks and subjective measures can be misleading in evaluating explainable AI systems. In Proceedings of the 25th International Conference on Intelligent User Interfaces . 454–464
Zana Buçinca, Phoebe Lin, Krzysztof Z Gajos, and Elena L Glassman. 2020 · 2020
Closest in time.
The Emerging Landscape of Explainable AI Planning and Decision Making
Original
Tathagata Chakraborti, Sarath Sreedharan, and Subbarao Kambhampati. 2020 · 2020
Closest in time.
Questioning the AI: Informing Design Practices for Explainable AI User Experiences. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems . ACM
Q Vera Liao, Daniel Gruen, and Sarah Miller. 2020 · 2020
Closest in time.
Explainability fact sheets: a framework for systematic assessment of explainable approaches. In Proceedings of the 2020 Conference on Fairness, Accountability, and Transparency . 56–67
Kacper Sokol and Peter Flach. 2020 · 2020
Closest in time.
Effect of Confidence and Explanation on Accuracy and Trust Calibration in AI-Assisted Decision Making. In Proceedings of the Conference on Fairness, Accountability, and Transparency . ACM
Yunfeng Zhang, Q Vera Liao, and Rachel KE Bellamy. 2020 · 2020
Closest in time.

Beyond the bibliography

alphaXiv searches the wider corpus for related work and actual follow-ups.

Open on alphaXiv

alphaXiv is searching for related work…

Reading the bibliography…