Adapting a kidney exchange algorithm to align with human values

Abstract The efficient and fair allocation of limited resources is a classical problem in economics and computer science. In kidney exchanges, a central market maker allocates living kidney donors to patients in need of an organ. Patients and donors in kidney exchanges are prioritized using ad-hoc weights decided on by committee and then fed into an allocation algorithm that determines who gets what—and who does not. In this paper, we provide an end-to-end methodology for estimating weights of individual participant profiles in a kidney exchange. We first elicit from human subjects a list of patient attributes they consider acceptable for the purpose of prioritizing patients (e.g., medical characteristics, lifestyle choices, and so on). Then, we ask subjects comparison queries between patient profiles and estimate weights in a principled way from their responses. We show how to use these weights in kidney exchange market clearing algorithms. We then evaluate the impact of the weights in simulations and find that the precise numerical values of the weights we computed matter little, other than the ordering of profiles that they imply. However, compared to not prioritizing patients at all, there is a significant effect, with certain classes of patients being (de)prioritized based on the human-elicited value judgments.

[1]  Jure Leskovec,et al.  Human Decisions and Machine Predictions , 2017, The quarterly journal of economics.

[2]  Guillermo Ricardo Simari,et al.  Argumentation in Artificial Intelligence , 2009 .

[3]  David Manlove,et al.  Position-Indexed Formulations for Kidney Exchange , 2016, EC.

[4]  M. Utku Ünver,et al.  A Kidney Exchange Clearinghouse in New England. , 2005, The American economic review.

[5]  D. Gamarnik,et al.  The Need for (Long) Chains in Kidney Exchange , 2012 .

[6]  Ralph A. Bradley,et al.  14 Paired comparisons: Some basic procedures and examples , 1984, Nonparametric Methods.

[7]  Tuomas Sandholm,et al.  Operation Frames and Clubs in Kidney Exchange , 2017, IJCAI.

[8]  C. Allen,et al.  Artificial Morality: Top-down, Bottom-up, and Hybrid Approaches , 2005, Ethics and Information Technology.

[9]  A. M. Turing,et al.  Computing Machinery and Intelligence , 1950, The Philosophy of Artificial Intelligence.

[10]  Luís Moniz Pereira,et al.  Agent Morality via Counterfactuals in Logic Programming , 2017, Bridging@CogSci.

[11]  Ruosong Wang,et al.  Efficient Near-optimal Algorithms for Barter Exchange , 2017, AAMAS.

[12]  David C. Parkes,et al.  Design and analysis of multi-hospital kidney exchange mechanisms using random graphs , 2015, Games Econ. Behav..

[13]  Francesca Rossi,et al.  Embedding Ethical Principles in Collective Decision Support Systems , 2016, AAAI.

[14]  Ariel D. Procaccia,et al.  Failure-Aware Kidney Exchange , 2019, Manag. Sci..

[15]  Martin W. P. Savelsbergh,et al.  Branch-and-Price: Column Generation for Solving Huge Integer Programs , 1998, Oper. Res..

[16]  Alvin E. Roth,et al.  Pairwise Kidney Exchange , 2004, J. Econ. Theory.

[17]  D. Segev,et al.  Domino paired kidney donation: a strategy to make best use of live non-directed donation , 2006, The Lancet.

[18]  Iyad Rahwan,et al.  The social dilemma of autonomous vehicles , 2015, Science.

[19]  Kristiaan M. Glorie,et al.  Kidney Exchange with Long Chains: An Efficient Pricing Algorithm for Clearing Barter Exchanges with Branch-and-Price , 2014, Manuf. Serv. Oper. Manag..

[20]  C. Allen,et al.  Moral Machines: Teaching Robots Right from Wrong , 2008 .

[21]  Matthias Scheutz,et al.  Contextual Deontic Cognitive Event Calculi for Ethically Correct Robots , 2018, ISAIM.

[22]  M. Utku Ünver,et al.  A nonsimultaneous, extended, altruistic-donor chain. , 2009, The New England journal of medicine.

[23]  Holger Hermanns,et al.  From Machine Ethics To Machine Explainability and Back , 2018, International Symposium on Artificial Intelligence and Mathematics.

[24]  Vincent Conitzer,et al.  Moral Decision Making Frameworks for Artificial Intelligence , 2017, ISAIM.

[25]  Toby Walsh,et al.  Mechanisms for Online Organ Matching , 2017, IJCAI.

[26]  A. Roth,et al.  Free riding and participation in large scale, multi‐hospital kidney exchange , 2014 .

[27]  Cathy O'Neil,et al.  Weapons of Math Destruction: How Big Data Increases Inequality and Threatens Democracy , 2016 .

[28]  Jianyang Zeng,et al.  Approximation of barter exchanges with cycle length constraints , 2016, ArXiv.

[29]  Edith Elkind,et al.  Rationalizations of Voting Rules , 2016, Handbook of Computational Social Choice.

[30]  Michael Anderson,et al.  Machine Ethics , 2011 .

[31]  Jian Li,et al.  Egalitarian pairwise kidney exchange: fast algorithms vialinear programming and parametric flow , 2014, AAMAS.

[32]  Péter Biró,et al.  Inapproximability of the kidney exchange problem , 2007, Inf. Process. Lett..

[33]  Tuomas Sandholm,et al.  FutureMatch: Combining Human Value Judgments and Machine Learning to Match in Dynamic Environments , 2015, AAAI.

[34]  David Manlove,et al.  Maximum Weight Cycle Packing in Directed Graphs, with Application to Kidney Exchange Programs , 2009, Discret. Math. Algorithms Appl..

[35]  Iyad Rahwan,et al.  A Voting-Based System for Ethical Decision Making , 2017, AAAI.

[36]  Avrim Blum,et al.  Clearing algorithms for barter exchange markets: enabling nationwide kidney exchanges , 2007, EC '07.

[37]  Ulises Cortés,et al.  Deliberation dialogues for reasoning about safety critical actions , 2012, Autonomous Agents and Multi-Agent Systems.

[38]  D. Gamarnik,et al.  Finding long chains in kidney exchange using the traveling salesman problem , 2015, Proceedings of the National Academy of Sciences.

[39]  David Manlove,et al.  Paired and Altruistic Kidney Donation in the UK , 2015, ACM J. Exp. Algorithmics.

[40]  David Sarne,et al.  Strategy-Proof and Efficient Kidney Exchange Using a Credit Mechanism , 2015, AAAI.

[41]  Ariel D. Procaccia,et al.  Opting Into Optimal Matchings , 2016, SODA.