Assessing Risks of Biases in Cognitive Decision Support Systems

Recognizing, assessing, countering, and mitigating the biases of different nature from heterogeneous sources is a critical problem in designing a cognitive Decision Support System (DSS). An example of such a system is a cognitive biometric-enabled security checkpoint. Biased algorithms affect the decision-making process in an unpredictable way, e.g. face recognition for different demographic groups may severely impact the risk assessment at a checkpoint. This paper addresses a challenging research question on how to manage an ensemble of biases? We provide performance projections of the DSS operational landscape in terms of biases. A probabilistic reasoning technique is used for assessment of the risk of such biases. We also provide a motivational experiment using face biometric component of the checkpoint system which highlights the discovery of an ensemble of biases and the techniques to assess their risks.

[1]  Catherine M. Burns,et al.  Intelligent Adaptive Systems: An Interaction-Centered Design Perspective , 2014 .

[2]  Feng Xu,et al.  Multi-Aspect + Transitivity + Bias: An Integral Trust Inference Model , 2014, IEEE Transactions on Knowledge and Data Engineering.

[3]  D. Winterfeldt,et al.  Cognitive and Motivational Biases in Decision and Risk Analysis , 2015, Risk analysis : an official publication of the Society for Risk Analysis.

[4]  Ming Hou,et al.  Intelligent Adaptive Interfaces for the Control of Multiple UAVs , 2007 .

[5]  Patrick Loiseau,et al.  Identity vs. Attribute Disclosure Risks for Users with Multiple Social Profiles , 2017, ASONAM.

[6]  Myles D. Garvey,et al.  An analytical framework for supply network risk propagation: A Bayesian network approach , 2015, Eur. J. Oper. Res..

[7]  Kurt Hugenberg,et al.  Towards a synthetic model of own group biases in face memory , 2013 .

[8]  Minqiang Li,et al.  A security risk analysis model for information systems: Causal relationships of risk factors and vulnerability propagation analysis , 2014, Inf. Sci..

[9]  Manoj Kumar Tiwari,et al.  Bayesian network modelling for supply chain risk propagation , 2018, Int. J. Prod. Res..

[10]  John R. Smith,et al.  Diversity in Faces , 2019, ArXiv.

[11]  Neera Jain,et al.  Computational Modeling of the Dynamics of Human Trust During Human–Machine Interactions , 2019, IEEE Transactions on Human-Machine Systems.

[12]  Joint Task Force.  Security and Privacy Controls for Information Systems and Organizations , 2020 .

[13]  Antitza Dantcheva,et al.  Mitigating Bias in Gender, Age and Ethnicity Classification: A Multi-task Convolution Neural Network Approach , 2018, ECCV Workshops.

[14]  Svetlana Yanushkevich,et al.  Cognitive checkpoint: Emerging technologies for biometric-enabled watchlist screening , 2019, Comput. Secur..

[15]  Harry Wechsler,et al.  The FERET database and evaluation procedure for face-recognition algorithms , 1998, Image Vis. Comput..

[16]  Keeley A. Crockett,et al.  Cognitive Identity Management: Risks, Trust and Decisions using Heterogeneous Sources , 2019, 2019 IEEE First International Conference on Cognitive Machine Intelligence (CogMI).

[17]  Jennifer Golbeck,et al.  SUNNY: A New Algorithm for Trust Inference in Social Networks Using Probabilistic Confidence Models , 2007, AAAI.

[18]  Lisa Hartling,et al.  Technology-assisted risk of bias assessment in systematic reviews: a prospective cross-sectional evaluation of the RobotReviewer machine learning tool. , 2017, Journal of clinical epidemiology.

[19]  Shawn Eastwood,et al.  Bridging the Gap Between Forensics and Biometric-Enabled Watchlists for e-Borders , 2017, IEEE Computational Intelligence Magazine.

[20]  Richong Zhang,et al.  Trust Prediction via Belief Propagation , 2014, ACM Trans. Inf. Syst..