Machine Assisted Human Decision Making

Artificial intelligence is often touted as the ultimate automation technology capable of outperforming humans. It is also feared by some because of its potential to eliminate certain jobs. In this paper, we describe scenarios in which man-machine symbiosis, or properly designed combinations of man and machine, can actually outperform man and machine. We also present a statistical solution to a constrained version of a generic problem in man-machine symbiosis. Specifically, we solve the problem of optimal selection, ordering and presentation of data to a human to solve a class of problems that artificial intelligence can fail to solve on its own, such as fraud detection. The man-machine symbiosis solution we present overcomes human cognitive biases which stand in the way of their rational decision making.

[1]  Michael H. Birnbaum,et al.  Bayesian Inference : Combining Base Rates With Opinions of Sources Who Vary in Credibility , 2005 .

[2]  Ahmed H. Tewfik,et al.  Cognitive biases in Bayesian updating and optimal information sequencing , 2015, 2015 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[3]  Ahmed H. Tewfik,et al.  Optimal information ordering for sequential detection with cognitive biases , 2016, 2016 24th European Signal Processing Conference (EUSIPCO).

[4]  L. Ross,et al.  Human Inference: Strategies and Shortcomings of Social Judgment. , 1981 .

[5]  Jason F. Shogren,et al.  Starting Point Bias in Dichotomous Choice Valuation with Follow-Up Questioning , 1996 .

[6]  E. Michael Bamber,et al.  An examination of the descriptive validity of the belief-adjustment model and alternative attitudes to evidence in auditing , 1997 .

[7]  Mark Levene.  Chess Metaphors, Artificial Intelligence and the Human Mind , 2011, Comput. J..

[8]  Xiaojin Zhu,et al.  Cognitive Models of Test-Item Effects in Human Category Learning , 2010, ICML.

[9]  Percy Liang,et al.  Understanding Black-box Predictions via Influence Functions , 2017, ICML.

[10]  Carey K. Morewedge,et al.  The Use of Theory in Designing a Serious Game for the Reduction of Cognitive Biases , 2016, DiGRA Conference.

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

[12]  R. Hogarth,et al.  Order effects in belief updating: The belief-adjustment model , 1992, Cognitive Psychology.

[13]  Tomek Strzalkowski,et al.  Effective Mitigation of Anchoring Bias, Projection Bias, and Representativeness Bias from Serious Game-based Training , 2015 .

[14]  Eli Upfal,et al.  Probability and Computing: Randomized Algorithms and Probabilistic Analysis , 2005 .

[15]  P. Weller,et al.  Quantifying Cognitive Biases in Analyst Earnings Forecasts , 2002 .

[16]  Ahmed H. Tewfik,et al.  Optimal information ordering in sequential detection problems with cognitive biases , 2014, 2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[17]  Erik M. Altmann.  Memory in Chains: Modeling Primacy and Recency Effects in Memory for Order , 2000 .

[18]  Bernard C. Levy,et al.  Principles of Signal Detection and Parameter Estimation , 2008 .

[19]  Matthew L. Jensen,et al.  Mitigating Cognitive Bias through the Use of Serious Games: Effects of Feedback , 2014, PERSUASIVE.

[20]  Cliff C. J. Huang,et al.  A general model of starting point bias in double-bounded dichotomous contingent valuation surveys , 2005 .

[21]  Carey K. Morewedge,et al.  Debiasing Decisions , 2015 .

[22]  D. Kahneman,et al.  Heuristics and Biases: The Psychology of Intuitive Judgment , 2002 .

[23]  Ahmed H. Tewfik,et al.  Real-time data selection and ordering for cognitive bias mitigation , 2016, 2016 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).