Accountability for Practical Reasoning Agents

Artificial intelligence has been increasing the autonomy of man-made artefacts such as software agents, self-driving vehicles and military drones. This increase in autonomy together with the ubiquity and impact of such artefacts in our daily lives have raised many concerns in society. Initiatives such as transparent and ethical AI aim to allay fears of a “free for all” future where amoral technology (or technology amorally designed) will replace humans with terrible consequences. We discuss the notion of accountable autonomy, and explore this concept within the context of practical reasoning agents. We survey literature from distinct fields such as management, healthcare, policy-making, and others, and differentiate and relate concepts connected to accountability. We present a list of justified requirements for accountable software agents and discuss research questions stemming from these requirements. We also propose a preliminary formalisation of one core aspect of accountability: responsibility.

[1]  Pablo Noriega,et al.  Normative Multi-Agent Systems , 2013, Normative Multi-Agent Systems.

[2]  Rosja Mastop,et al.  Characterising Responsibility in Organisational Structures: The Problem of Many Hands , 2010, DEON.

[3]  Hans Weigand,et al.  Meeting the Deadline: On the Formal Specification of Temporal Deontic Constraints , 1996, ISMIS.

[4]  James Zou,et al.  AI can be sexist and racist — it’s time to make it fair , 2018, Nature.

[5]  M. Bovens.  Analysing and Assessing Accountability: A Conceptual Framework , 2007 .

[6]  Peter Berliner,et al.  DEVELOPMENT OF PRACTICE , 2007 .

[7]  Joseph Y. Halpern,et al.  Causality, Responsibility and Blame in Team Plans , 2017, AAMAS.

[8]  Weiru Liu,et al.  Plan Selection for Probabilistic BDI Agents , 2014, 2014 IEEE 26th International Conference on Tools with Artificial Intelligence.

[9]  Emiliano Lorini,et al.  A logical analysis of responsibility attribution: emotions, individuals and collectives , 2014, J. Log. Comput..

[10]  Mehdi Dastani,et al.  Coalitional Responsibility in Strategic Settings , 2013, CLIMA.

[11]  Vincent P. Wade,et al.  The evaluation of adaptive and personalised information retrieval systems: a review , 2011, Int. J. Knowl. Web Intell..

[12]  Frank Dignum,et al.  Collective Obligations and Agents: Who Gets the Blame? , 2004, DEON.

[13]  J. Mike Jacka,et al.  Business Process Mapping: Improving Customer Satisfaction , 2002 .

[14]  Alain Finkel,et al.  Well-abstracted transition systems: application to FIFO automata , 2003, Inf. Comput..

[15]  Trevor J. M. Bench-Capon,et al.  Practical reasoning as presumptive argumentation using action based alternating transition systems , 2007, Artif. Intell..

[16]  Donald Perlis,et al.  Logic, Self-awareness and Self-improvement: the Metacognitive Loop and the Problem of Brittleness , 2005, J. Log. Comput..

[17]  Nico Roos,et al.  Primary and secondary diagnosis of multi-agent plan execution , 2009, Autonomous Agents and Multi-Agent Systems.

[18]  Nico Roos,et al.  Diagnosis of single and multi-agent plans , 2005, AAMAS '05.

[19]  Roberto Micalizio,et al.  The AThOS Project: First Steps towards Computational Accountability , 2017, CARe-MAS@PRIMA.

[20]  Frank Dignum,et al.  Organizational structure and responsibility , 2007, Artificial Intelligence and Law.

[21]  Roberto Micalizio,et al.  ADOPT JaCaMo: Accountability-Driven Organization Programming Technique for JaCaMo , 2017, PRIMA.

[22]  Michael Wooldridge,et al.  The Belief-Desire-Intention Model of Agency , 1998, ATAL.

[23]  Virginia Dignum,et al.  Ethics in artificial intelligence: introduction to the special issue , 2018, Ethics and Information Technology.

[24]  Munindar P. Singh,et al.  Accountability as a Foundation for Requirements in Sociotechnical Systems , 2014, IEEE Internet Computing.

[25]  Wamberto Weber Vasconcelos,et al.  Group Norms for Multi-Agent Organisations , 2016, ACM Trans. Auton. Adapt. Syst..

[26]  Marco Colombetti,et al.  Representation and monitoring of commitments and norms using OWL , 2010, AI Commun..

[27]  Michael Winikoff.  An AgentSpeak Meta-interpreter and Its Applications , 2005, PROMAS.

[28]  Mehdi Dastani,et al.  Commitments and interaction norms in organisations , 2015, Autonomous Agents and Multi-Agent Systems.

[29]  Michael Winikoff,et al.  No Pizza for You: Value-based Plan Selection in BDI Agents , 2017, IJCAI.

[30]  Munindar P. Singh,et al.  An algebra for commitment protocols , 2007, Autonomous Agents and Multi-Agent Systems.

[31]  Michael Luck,et al.  BDI reasoning with normative considerations , 2015, Eng. Appl. Artif. Intell..

[32]  Michael Winikoff,et al.  Debugging Agent Programs with Why?: Questions , 2017, AAMAS.

[33]  Pietro Torasso,et al.  On-Line Monitoring and Diagnosis of Multi-Agent Systems: A Model Based Approach , 2004, ECAI.

[34]  J. Fox,et al.  The uncertain relationship between transparency and accountability , 2007 .

[35]  L. Emanuel,et al.  What Is Accountability in Health Care? , 1996, Annals of Internal Medicine.

[36]  Pietro Torasso,et al.  Cooperative Monitoring to Diagnose Multiagent Plans , 2014, J. Artif. Intell. Res..

[37]  Avelino Francisco Zorzo,et al.  Propositional planning in BDI agents , 2004, SAC '04.

[38]  R. Mulgan.  'Accountability': an ever-expanding concept? , 2000 .

[39]  Michael Winikoff,et al.  Towards Trusting Autonomous Systems , 2017, EMAS@AAMAS.

[40]  T. Schillemans.  The Public Accountability Review. a Meta-Analysis of Public Accountability Research in Six Academic Disciplines , 2013 .

[41]  Melvin J. Dubnick.  Accountability as a Cultural Keyword , 2014 .

[42]  R. Goodin,et al.  The Oxford Handbook of Public Accountability , 2014 .

[43]  Anand S. Rao,et al.  BDI Agents: From Theory to Practice , 1995, ICMAS.

[44]  Frank Dignum,et al.  Modeling the problem of many hands in organisations , 2010, ECAI.

[45]  Eliezer Yudkowsky,et al.  The Ethics of Artificial Intelligence , 2014, Artificial Intelligence Safety and Security.

[46]  Mehdi Dastani,et al.  Reasoning about plan revision in BDI agent programs , 2011, Theor. Comput. Sci..

[47]  Albert Gatt,et al.  From data to text in the Neonatal Intensive Care Unit: Using NLG technology for decision support and information management , 2009, AI Commun..

[48]  Edgar A. Whitley,et al.  The Construction of Social Reality , 1999 .

[49]  Virginia Dignum,et al.  Responsible Artificial Intelligence: Designing Ai for Human Values , 2017 .