W.F. Lawless · Ranjeev Mittu Donald Sofge · Stephen Russell Editors Autonomy and Artifi cial Intelligence: A Threat or Savior? Autonomy and Artificial Intelligence: A Threat or Savior? W.F. Lawless • Ranjeev Mittu • Donald Sofge Stephen Russell Editors Autonomy and Artificial Intelligence: A Threat or Savior? Editors W.F. Lawless Ranjeev Mittu Paine College Naval Research Laboratory Augusta, GA, USA Washington, DC, USA Donald Sofge Stephen Russell Naval Research Laboratory U.S. Army Research Laboratory Washington, DC, USA Adelphi, MD, USA ISBN 978-3-319-59718-8 ISBN 978-3-319-59719-5 (eBook) DOI 10.1007/978-3-319-59719-5 Library of Congress Control Number: 2017947297 © Springer International Publishing AG 2017 Chapters 1, 4, 5, 12, and 13 were created within the capacity of an US governmental employment. US copyright protection does not apply. This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, express or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Printed on acid-free paper This Springer imprint is published by Springer Nature The registered company is Springer International Publishing AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland Preface This book derives from two Association for the Advancement of Artificial Intelligence (AAAI) symposia; the first symposium on “Foundations of Autonomy and Its (Cyber) Threats—From Individuals to Interdependence” was held at Stanford University from March 23 to 25, 2015, and the second symposium on “AI and the Mitigation of Human Error—Anomalies, Team Metrics and Thermodynamics” was held again at Stanford University from March 21 to 23, 2016. This book, titled Autonomy and Artificial Intelligence: A Threat or Savior?, combines and extends the themes of both symposia. Our goal for this book is to deal with the current state of the art in autonomy and artificial intelligence by examining the gaps in the exist- ing research that must be addressed to better integrate autonomous and human sys- tems. The research we present in this book will help to advance the next generation of systems that are already planned ranging from autonomous platforms and machines to teams of autonomous systems to provide better support to human oper- ators, decision-makers, and the society. This book explores how artificial intelligence (AI), by leading to an increase in the autonomy of machines and robots, is offering opportunities for an expanded but uncertain impact on society by humans, machines, and robots. To help readers better understand the relationships between AI, autonomy, humans, and machines that will help society reduce human errors in the use of advanced technologies (e.g., air- planes, trains, cars), this edited volume presents a wide selection of the underlying theories, computational models, experimental methods, and field applications. While other books deal with these topics individually, this book is unique in that it unifies the fields of autonomy and AI and frames them in the broader context of effective integration for human-autonomous machine and robotic systems. The introduction in this volume begins by describing the current state of the art for research in AI, autonomy, and cyber-threats presented at Stanford University in the spring of 2015 (copies of the technical articles are available from AAAI at http:// www.aaai.org/Symposia/Spring/sss15symposia.php#ss03; a link to the agenda for the symposium in 2015 along with contact information for the invited speakers and regular participants is at https://sites.google.com/site/foundationsofautonomy- aaais2015/) and for research in AI, autonomy, and error mitigation presented at the v vi Preface same university in the spring of 2016 (copies of the technical articles are available from AAAI at http://www.aaai.org/Symposia/Spring/sss16symposia.php#ss01; a link to the agenda and contact information for the invited speakers and regular par- ticipants is at https://sites.google.com/site/aiandthemitigationofhumanerror/). After introducing the themes in this book and the contributions from world-class researchers and scientists, individual chapters follow where they elaborate on key research topics at the heart of effective human-machine-robot-systems integration. These topics include computational support for intelligence analyses; the challenge of verifying today’s and future autonomous systems; comparisons between today’s machines and autism; implications of human-information interaction on artificial intelligence and errors; systems that reason; the autonomy of machines, robots, and buildings; and hybrid teams, where hybrid reflects arbitrary combinations of humans, machines, and robots. The contributions to this volume are written by leading scientists across the field of autonomous systems research, ranging from industry and academia to govern- ment. Given the broad diversity of the research in this book, we strove to thoroughly examine the challenges and trends of systems that implement and exhibit AI; social implications of present and future systems made autonomous with AI; systems with AI seeking to develop trusted relationships among humans, machines, and robots; and effective human systems integration that must result for trust in these new sys- tems and their applications to increase and to be sustained. A brief summary of the AAAI symposia in the spring of 2015 and the spring of 2016 is presented below. Spring 2015: Foundations of Autonomy and Its (Cyber) Threats—From Individuals to Interdependence Spring 2015: Organizing Committee Ranjeev Mittu ([email protected]), Naval Research Laboratory Gavin Taylor ([email protected]), US Naval Academy Donald Sofge ([email protected]), Naval Research Laboratory, Navy Center for Applied Research in Artificial Intelligence William F. Lawless ([email protected]), Paine College, Departments of Math and Psychology Spring 2015: Program Committee • David Atkinson ([email protected]), Senior Research Scientist, Florida Institute for Human and Machine Cognition Preface vii • Lashon B. Booker ([email protected]), Ph.D., Senior Principal Scientist, The MITRE Corporation • Jeffery Bradshaw ([email protected]), Senior Research Scientist, Florida Institute for Human and Machine Cognition • Michael Floyd ([email protected]), Knexus Research • Sharon Graves ([email protected]), NASA Deputy Project Manager, Safe Autonomous Systems Operations, Aeronautics Research Directorate • Vladimir Gontar ([email protected]), Department of Industrial Engineering and Management, Ben-Gurion University of the Negev • L. Magafas ([email protected]), Director of Electronics and Signal Processing Lab., Eastern Macedonia and Thrace Institute of Technology, Kavala, GR • Bolivar Rocha ([email protected]), Brazil • Satyandra K. Gupta ([email protected]), Director, University of Maryland Robotics Center, Department of Mechanical Engineering and Institute for Systems Research • Laurent Chaudron ([email protected]), Director, ONERA Provence Research Center, French Air Force Academy • Charles Howell ([email protected]), Chief Engineer for Intelligence Programs and Integration, National Security Engineering Center, The MITRE Corporation • Jennifer Burke ([email protected]), Manager, Human-System Integrated Technologies, Boeing Research and Technology • Tsuyoshi Murata ([email protected]), Dept. of Computer Science, Graduate School of Information Science and Engineering, Tokyo Institute of Technology • Julie Marble ([email protected]), Office of Naval Research, Program Officer for Hybrid Human-Computer Systems • Doug Riecken ([email protected]), Columbia University Center for Computational Learning Systems • Catherine Tessier ([email protected]), Senior Researcher, Dept. of Systems Control and Flight Dynamics, French Aerospace Lab, ONERA, Toulouse, France • Simon Parsons ([email protected]), Liverpool, Visiting Professor, Dept. of Computer Science, University of Liverpool; Dept. Graduate Deputy Chair and Co-Dir., Agents Lab, Brooklyn College • Ciara Sibley ([email protected]), Engineering Research Psychologist, Naval Research Laboratory, Washington, DC Spring 2015: Invited Keynote Speakers • Gautam Trivedi ([email protected]) and Brandon Enochs (brandon. [email protected]), Naval Research Laboratory, “Detecting, Analyzing and Locating Unauthorized Wireless Intrusions into Networks” • Chris Berka ([email protected]), Advanced Brain Monitoring, “On the Road to Autonomy: Evaluating and Optimizing Hybrid Team Dynamics” viii Preface • Kristin E. Schaefer ([email protected]), US Army Research Lab (ARL), “Perspectives of Trust: Research at the US Army Research Laboratory” • David R. Martinez ([email protected]), Lincoln Laboratory, Massachusetts Institute of Technology, “Cyber Anomaly Detection with Machine Learning” • Vladimir Gontar ([email protected]), BioCircuits Institute, University of California San Diego (UCSD), Ben-Gurion University of the Negev, “Artificial Brain Systems Based on Neural Networks Discrete Chaotic Biochemical Reactions Dynamics and Its Application to Conscious and Creative Robots” Spring 2015: Regular Speakers • Christopher A. Miller ([email protected]), Smart Information Flow Technologies, “Delegation, Intent, Cooperation and Their Failures” • Ciara Sibley1 ([email protected]), Joseph Coyne1 (joseph.coyne@nrl. navy.mil), and Jeffery Morrison2 ([email protected]), 1Naval Research Laboratory, 2Office of Naval Research, “Research Considerations for Managing Future Unmanned Systems” • Gavin Taylor ([email protected]), Kawika Barabin, and Kent Sayre, Computer Science Department, US Naval Academy, Annapolis, MD 21402-5002, “An Application of Reinforcement Learning to Supervised Autonomy” • David J.  Atkinson ([email protected]), Florida Institute for Human and Machine Cognition, Ocala, FL, “Emerging Cyber-Security Issues of Autonomy and the Psychopathology of Intelligent Machines” • Olivier Bartheye1 ([email protected]) and Laurent Chaudron2 ([email protected]), CREC St-Cyr1 and ONERA2, “Risk Management Systems Must Provide Automatic Decisions for Crisis Computable Algebras” • William F. Lawless ([email protected]), Paine College, Augusta, GA, and Ira S. Moskowitz, Ranjeev Mittu, and Donald A. Sofge ({ira.moskowitz; ranjeev. mittu; donald.sofge}@nrl.navy.mil), Naval Research Laboratory, Washington, DC, “A Thermodynamics of Teams: Towards a Robust Computational Model of Autonomous Teams” • Ranjeev Mittu1 ([email protected]) and Julie Marble2 (julie.marble@ navy.mil), 1Naval Research Laboratory, Information Technology Division, Washington, DC; 2 Office of Naval Research, VA 22203-1995 (changing to Johns Hopkins Applied Physics Lab, MD), “The Human Factor in Cybersecurity: Robust and Intelligent Defense” • Myriam Abramson ([email protected]), Naval Research Laboratory, Washington, DC, “Cognitive Fingerprints” • Ira S. Moskowitz1 ([email protected]), William F. Lawless2, (wlaw- [email protected]), Paul Hyden1 ([email protected]), Ranjeev Mittu1 (ran- Preface ix [email protected]), and Stephen Russell1 (stephen.m.russell8.civ@mail. mil), 1Information Management and Decision Architectures Branch, Naval Research Laboratory, Washington, DC; 2Departments of Mathematics and Psychology, Paine College, Augusta, GA, “A Network Science Approach to Entropy and Training” • Boris Galitsky ([email protected]), Knowledge Trail Inc., San Jose, CA, “Team Formation by Children with Autism” • Olivier Bartheye1 ([email protected]) and Laurent Chaudron2 ([email protected]), CREC St-Cyr1 and ONERA2, “Algebraic Models of the Self-Orientation Concept for Autonomous Systems” Spring 2016: AI and the Mitigation of Human Error— Anomalies, Team Metrics and Thermodynamics Spring 2016: Organizing Committee Ranjeev Mittu ([email protected]), Naval Research Laboratory Gavin Taylor ([email protected]), US Naval Academy Donald Sofge ([email protected]), Naval Research Laboratory William F. Lawless ([email protected]), Paine College, Departments of Math and Psychology Spring 2016: Program Committee (duplicates the spring 2015 symposium) Spring 2016: Invited Keynote Speakers • Julie Adams ([email protected]), Vanderbilt University, Associate Professor of Computer Science and Computer Engineering, Electrical Engineering and Computer Science Department, “AI and the Mitigation of Error” • Stephen Russell ([email protected]), Chief, Battlefield Information Processing Branch, US Army Research Lab, MD, “Human Information Interaction, Artificial Intelligence, and Errors” • James Llinas ([email protected]), SUNY at Buffalo, “An Argumentation-Based System Support Toolkit for Intelligence Analyses” • Martin Voshell ([email protected]), Charles River Analytics, “Multi-Level Human-Autonomy Teams for Distributed Mission Management” x Preface Spring 2016: Regular Speakers • Ira S.  Moskowitz ([email protected]), NRL; “Human-Caused Bifurcations in a Hybrid Team—A Position Paper” • Paul Hyden ([email protected]), NRL, “Fortification Through Topological Dominance: Using Hop Distance and Randomized Topology Strategies to Enhance Network Security” • Olivier Bartheye ([email protected]), CREC St-Cyr, , and Laurent Chaudron ([email protected]), ONERA, “Epistemological Qualification of Valid Action Plans for UGVs or UAVs in Urban Areas” • William F.  Lawless, ([email protected]), Paine College, “AI and the Mitigation of Error: A Thermodynamics of Teams” Questions for Speakers and Attendees at AAAI-2015 and AAAI-2016 and for Readers of This Book Our spring AAAI-2015 and AAAI-2016 symposia offered speakers opportunities with AI to address the intractable, fundamental questions about cybersecurity, machines and robots, autonomy and its management, the malleability of preferences and beliefs in social settings, or the application of autonomy for hybrids at the indi- vidual, group, and system levels. A list of unanswered fundamental questions included: • Why have we yet to determine from a theoretical perspective the principles underlying individual, team, and system behaviors? • Can autonomous systems be controlled to solve the problems faced by teams while maintaining defenses against threats and minimizing mistakes in competi- tive environments (e.g., cyber attacks, human error, system failure)? • Do individuals seek to self-organize into autonomous groups like teams in order to better defend against attacks (e.g., cyber, merger, resources) or for other rea- sons (e.g., least entropy production (LEP) and maximum entropy production (MEP))? • What does an autonomous organization need to predict its path forward and gov- ern itself? What are the AI tools available to help an organization be more adept and creative? • What signifies adaptation? For AI, does adaptation at an earlier time prevent or moderate adaptive responses to newer environmental changes? • Is the stability state of hybrid teams the single state that generates the MEP rate? • If social order requires MEP, and if the bistable perspectives present in debate (courtrooms, politics, science) lead to stable decisions, is the chosen decision an LEP or MEP state?