Machine Guides, Human Supervises: Interactive Learning with Global Explanations

We introduce explanatory guided learning (XGL), a novel interactive learning strategy in which a machine guides a human supervisor toward selecting informative examples for a classifier. The guidance is provided by means of global explanations, which summarize the classifier's behavior on different regions of the instance space and expose its flaws. Compared to other explanatory interactive learning strategies, which are machine-initiated and rely on local explanations, XGL is designed to be robust against cases in which the explanations supplied by the machine oversell the classifier's quality. Moreover, XGL leverages global explanations to open up the black-box of human-initiated interaction, enabling supervisors to select informative examples that challenge the learned model. By drawing a link to interactive machine teaching, we show theoretically that global explanations are a viable approach for guiding supervisors. Our simulations show that explanatory guided learning avoids overselling the model's quality and performs comparably or better than machine- and human-initiated interactive learning strategies in terms of model quality.

[1]  Jude W. Shavlik,et al.  in Advances in Neural Information Processing , 1996 .

[2]  Panagiotis Papapetrou,et al.  A peek into the black box: exploring classifiers by randomization , 2014, Data Mining and Knowledge Discovery.

[3]  Xiaojin Zhu,et al.  Machine Teaching: An Inverse Problem to Machine Learning and an Approach Toward Optimal Education , 2015, AAAI.

[4]  Albert Gordo,et al.  Learning Global Additive Explanations for Neural Nets Using Model Distillation , 2018 .

[5]  Thomas G. Dietterich.  Steps Toward Robust Artificial Intelligence , 2017, AI Mag..

[6]  David Maxwell Chickering,et al.  ICE: Enabling Non-Experts to Build Models Interactively for Large-Scale Lopsided Problems , 2014, ArXiv.

[7]  Kristian Kersting,et al.  Right for the Wrong Scientific Reasons: Revising Deep Networks by Interacting with their Explanations , 2020, ArXiv.

[8]  Samuel J. Gershman,et al.  Human-in-the-Loop Interpretability Prior , 2018, NeurIPS.

[9]  Kristian Kersting,et al.  Explanatory Interactive Machine Learning , 2019, AIES.

[10]  Panagiotis G. Ipeirotis,et al.  Beat the Machine: Challenging Humans to Find a Predictive Model's “Unknown Unknowns” , 2015, JDIQ.

[11]  João Gama,et al.  A survey on concept drift adaptation , 2014, ACM Comput. Surv..

[12]  Bin Li,et al.  A survey on instance selection for active learning , 2012, Knowledge and Information Systems.

[13]  Albert Gordo,et al.  Transparent Model Distillation , 2018, ArXiv.

[14]  Lars Niklasson,et al.  Accuracy vs. comprehensibility in data mining models , 2004 .

[15]  Edith Law,et al.  Contradict the Machine: A Hybrid Approach to Identifying Unknown Unknowns , 2019, AAMAS.

[16]  Carlos Guestrin,et al.  "Why Should I Trust You?": Explaining the Predictions of Any Classifier , 2016, ArXiv.

[17]  Maya Cakmak,et al.  Eliciting good teaching from humans for machine learners , 2014, Artif. Intell..

[18]  Pietro Perona,et al.  Near-Optimal Machine Teaching via Explanatory Teaching Sets , 2018, AISTATS.

[19]  Stefano Teso.  Toward Faithful Explanatory Active Learning with Self-explainable Neural Nets , 2019 .

[20]  Osbert Bastani,et al.  Interpreting Blackbox Models via Model Extraction , 2017, ArXiv.

[21]  Sanjay Ranka,et al.  Global Model Interpretation Via Recursive Partitioning , 2018, 2018 IEEE 20th International Conference on High Performance Computing and Communications; IEEE 16th International Conference on Smart City; IEEE 4th International Conference on Data Science and Systems (HPCC/SmartCity/DSS).

[22]  Cynthia Rudin,et al.  Interpretable Image Recognition with Hierarchical Prototypes , 2019, HCOMP.

[23]  Sanjoy Dasgupta,et al.  Teaching a black-box learner , 2019, ICML.

[24]  Steve Hanneke,et al.  Theory of Disagreement-Based Active Learning , 2014, Found. Trends Mach. Learn..

[25]  Eric Horvitz,et al.  Identifying Unknown Unknowns in the Open World: Representations and Policies for Guided Exploration , 2016, AAAI.

[26]  Fausto Giunchiglia,et al.  Fixing Mislabeling by Human Annotators Leveraging Conflict Resolution and Prior Knowledge , 2019, Proc. ACM Interact. Mob. Wearable Ubiquitous Technol..

[27]  T. Kathirvalavakumar,et al.  Reverse Engineering the Neural Networks for Rule Extraction in Classification Problems , 2011, Neural Processing Letters.

[28]  Manuel Lopes,et al.  Interactive Optimal Teaching with Unknown Learners , 2018, IJCAI.

[29]  Terrance E. Boult,et al.  Learning and the Unknown: Surveying Steps toward Open World Recognition , 2019, AAAI.

[30]  Giles Hooker,et al.  Tree Space Prototypes: Another Look at Making Tree Ensembles Interpretable , 2016, FODS.

[31]  Zhiyuan Liu,et al.  Incentivized Exploration for Multi-Armed Bandits under Reward Drift , 2020, AAAI.

[32]  David Cohn,et al.  Active Learning , 2010, Encyclopedia of Machine Learning.

[33]  Franco Turini,et al.  A Survey of Methods for Explaining Black Box Models , 2018, ACM Comput. Surv..

[34]  John Langford,et al.  Search Improves Label for Active Learning , 2016, NIPS.

[35]  Mark Craven,et al.  Understanding Learned Models by Identifying Important Features at the Right Resolution , 2018, AAAI.

[36]  Rich Caruana,et al.  Model compression , 2006, KDD '06.

[37]  Eric Horvitz,et al.  Beyond Accuracy: The Role of Mental Models in Human-AI Team Performance , 2019, HCOMP.

[38]  Klaus-Robert Müller,et al.  Explanations can be manipulated and geometry is to blame , 2019, NeurIPS.

[39]  Olcay Boz,et al.  Extracting decision trees from trained neural networks , 2002, KDD.

[40]  Andrew P. Bradley,et al.  Rule extraction from support vector machines: A review , 2010, Neurocomputing.

[41]  Joachim Diederich,et al.  Survey and critique of techniques for extracting rules from trained artificial neural networks , 1995, Knowl. Based Syst..

[42]  Joachim Diederich,et al.  Rule Extraction from Support Vector Machines , 2008, Studies in Computational Intelligence.

[43]  Foster J. Provost,et al.  Why label when you can search?: alternatives to active learning for applying human resources to build classification models under extreme class imbalance , 2010, KDD.