Trusted Machine Learning for Probabilistic Models

In several mission-critical domains (e.g., selfdriving cars, cybersecurity, robotics) where machine learning algorithms are being used heavily, it is becoming increasingly important to ensure that the learned models satisfy some domain properties (e.g., temporal constraints). Towards this goal, we propose Trusted Machine Learning (TML), wherein we combine the strengths of machine learning and model checking. If the desired logical properties are not satisfied by a trained model, we modify either the model (‘model repair’) or the data from which the model is learned (‘data repair’). We outline a concrete case study based on the Markov Chain model of a car controller for ‘lane changing’ — we demonstrate how we can ensure that such a model, learned from data, satisfies properties specified in Probabilistic Computation Tree Logic (PCTL).

[1]  Jun Zhu,et al.  Robust RegBayes: Selectively Incorporating First-Order Logic Domain Knowledge into Bayesian Models , 2014, ICML.

[2]  Lijun Zhang,et al.  PARAM: A Model Checker for Parametric Markov Models , 2010, CAV.

[3]  S. Shankar Sastry,et al.  A learning based approach to control synthesis of Markov decision processes for linear temporal logic specifications , 2014, 53rd IEEE Conference on Decision and Control.

[4]  Alberto L. Sangiovanni-Vincentelli,et al.  Polynomial-Time Verification of PCTL Properties of MDPs with Convex Uncertainties , 2013, CAV.

[5]  C. R. Ramakrishnan,et al.  Model Repair for Probabilistic Systems , 2011, TACAS.

[6]  Orna Kupferman,et al.  Freedom, weakness, and determinism: from linear-time to branching-time , 1998, Proceedings. Thirteenth Annual IEEE Symposium on Logic in Computer Science (Cat. No.98CB36226).

[7]  Kenneth D. Miller,et al.  The Role of Constraints in Hebbian Learning , 1994, Neural Computation.

[8]  Tomer Hertz,et al.  Learning a Mahalanobis Metric from Equivalence Constraints , 2005, J. Mach. Learn. Res..

[9]  David J. Kriegman,et al.  Leveraging temporal, contextual and ordering constraints for recognizing complex activities in video , 2007, 2007 IEEE Conference on Computer Vision and Pattern Recognition.

[10]  Matthew Richardson,et al.  Markov logic networks , 2006, Machine Learning.

[11]  Xiaojin Zhu,et al.  Using Machine Teaching to Identify Optimal Training-Set Attacks on Machine Learners , 2015, AAAI.

[12]  Alessandro Sperduti,et al.  Online Process Discovery to Detect Concept Drifts in LTL-Based Declarative Process Models , 2013, OTM Conferences.

[13]  Alberto L. Sangiovanni-Vincentelli,et al.  Data-Driven Probabilistic Modeling and Verification of Human Driver Behavior , 2014, AAAI Spring Symposia.

[14]  Thomas Glen Dietterich Constraint propagation techniques for theory-driven data interpretation (artificial intelligence, machine learning) , 1985 .

[15]  D. Parker Verification of Probabilistic Real-time Systems , 2013 .

[16]  Wojciech Kotlowski,et al.  Rule learning with monotonicity constraints , 2009, ICML '09.

[17]  Richard E. Neapolitan,et al.  Learning Bayesian networks , 2007, KDD '07.

[18]  Marta Z. Kwiatkowska,et al.  PRISM 4.0: Verification of Probabilistic Real-Time Systems , 2011, CAV.

[19]  Luc De Raedt,et al.  Constraint Programming for Data Mining and Machine Learning , 2010, AAAI.

[20]  Dimitri P. Bertsekas,et al.  Constrained Optimization and Lagrange Multiplier Methods , 1982 .

[21]  S. Wallace Algorithms and Model Formulations in Mathematical Programming , 1989, NATO ASI Series.

[22]  Stein W. Wallace Algorithms and model formulations in mathematical programming , 1989 .

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

[24]  Barbara König,et al.  Tools and Algorithms for the Construction and Analysis of Systems , 2012, Lecture Notes in Computer Science.