Preface of the 2018 Symposium on Adversary Aware Learning Techniques and Trends in Cybersecurity (ALEC) (co-located with AAAI Fall Symposium Series 2018) Prithviraj Dasgupta, Joseph B. Collins, Ranjeev Mittu∗ Machine learning-based intelligent systems have experi- • Operations Research related to Adversarial Learning enced a massive growth over the past few years, and are • Applications of Adversarial Learning close to becoming ubiquitous in the technology surround- ing our daily lives. However, a critical challenge in machine • Security Threats and Vulnerabilities of Adversarial Learn- learning-based systems is their vulnerability to security at- ing tacks from malicious adversaries. The vulnerability of these • Human factors and adversarial learning with human-in- systems is further aggravated as it is non-trivial to establish the-loop the authenticity of data used to train the system, and even The symposium included two keynote talks and ten orally innocuous perturbations to the training data can be used to presented papers. The first keynote talk titled AI Canonical manipulate the systems behavior in unintended ways. Re- Architecture and Robust AI by David R. Martinez from MIT cent reports about malicious manipulation of social media Lincoln Laboratories discussed the performance assessment feeds masquerading as authentic news items provide com- of AI-based systems and the need for robust AI. Xiaojin pelling evidence towards developing stronger and more re- (Jerry) Zhu from the University of Wisconsin-Madison pre- silient measures for combating adversarial attacks on ma- sented the second keynote titled An Optimal Control View chine learning-based systems. of Adversarial Machine Learning on a novel control theory- The ALEC’18 symposium was organized to address based framework for representing various adversarial learn- the overarching need towards making automated, machine ing problems. The research papers presented at the sym- learning-based systems more robust and resilient against ad- posium were grouped into three theme-based sessions: (1) versarial attacks, so that humans can use them in a safe and Adversarial Data Generation and Adversarial Training, (2) sustained manner. The areas of interest of the symposium Countering Adversarial Attacks in Cybersecurity, and, (3) included the following topics: Novel Approaches in Adversarial Artificial Intelligence. The • Adversary-aware Machine Learning - Reinforcement symposium concluded with a group discussion on the imme- Learning, Lifelong Learning, Deep Learning diate technological enablers and hurdles in adversarial learn- ing as well as determining a roadmap for addressing longer • Generative Adversarial Networks term problems and challenges in the field. • Adversary- aware Prediction, Forecasting and Decision Finally, we would like to thank the following ALEC’18 Making Techniques program committee members and reviewers for their support • Game Theory and Game Playing to counter adversarial with reviewing papers and with various aspects of organiz- learning ing the symposium: • Distributed, Multi-agent Systems • Amitabh Mishra, U.S. Army CERDEC, USA • Adversarial Issues and Techniques for Cyber-Physical • Abebaw Tadesse, Langston University, USA Systems, Adversarial Robotics • Krishnendu Ghosh, Miami University of Ohio, USA ∗ P. Dasgupta is with the Computer Science Department, • Ying Zhao, Naval Postgraduate School, USA University of Nebraska, Omaha, NE, USA. e-mail: pdas- November 7, 2018. gupta@unomaha.edu. J. Collins and R. Mittu are with the U.S. Naval Research Laboratory, Washington D.C., USA. Email: Prithviraj Dasgupta {joseph.collins, ranjeev.mittu}@nrl.navy.mil Joseph B. Collins Copyright c by the paper’s authors. Copying permitted for private Ranjeev Mittu and academic purposes. In: Joseph Collins, Prithviraj Dasgupta, Ranjeev Mittu (eds.): Proceedings of the AAAI Fall 2018 Sympo- sium on Adversary-Aware Learning Techniques and Trends in Cy- bersecurity, Arlington, VA, USA, 18-19 October, 2018, published at http://ceur-ws.org