ACM Workshop on Wireless Security and Machine Learning (WiseML 2020)

The second ACM Workshop on Wireless Security and Machine Learning (WiseML 2020) will be held on July 13, 2020 in conjunction with the 13th ACM Conference on Security and Privacy in Wireless and Mobile Networks (ACM WiSec 2020). The workshop will be held virtually with online presentations. Accepted papers will appear in the conference proceedings and the ACM digital library.

Scope and background

Artificial intelligence (AI) and machine learning (ML) have proven to be invaluable tools for a diverse and far-reaching set of applications ranging from traditional image recognition and natural language processing tasks to more recent cyber security and autonomous navigation challenges. In recent years, applications of AI/ML have also emerged in the wireless communications and networking domains. ML systems based upon state-of-the-art neural network architectures, powered by the ever-more powerful hardware accelerators for computing, have been deployed for spectrum sensing applications (signal detection, estimation, classification, and identification), channel estimation and feedback, coding, waveform design, emitter identification, cognitive jamming and anti-jamming, among many others.

In the more established AI/ML domains, recent research has demonstrated the efficacy of adversarial ML (AML) techniques to negatively impact the performance of AI/ML systems with minimal distortion on the data samples. Consequently, vulnerabilities to the privacy and security of these systems, and the data used to train them, have been exposed. However, the impact of these concepts on wireless technologies is at present underdeveloped. In addition, the proliferation of mobile collaborative multi-agent autonomous platforms (aerial, terrestrial, or underwater), communicating over wireless networks, has made them susceptible targets to various attacks. Harnessing efficient, robust AI/ML algorithms for wireless security that can operate under constrained power and computational resources, is of paramount importance for guaranteeing the integrity of such platforms. Undoubtedly, an effort to investigate the interactions between ML and wireless security, privacy, and robustness, would be both timely and indispensable.

The purpose of this workshop is to bring together members of the AI/ML, privacy, security, wireless communications and networking communities from around the world and offer them the opportunity to share the latest research findings in these emerging and critical areas, as well as to exchange ideas and foster research collaborations, in order to further advance the state-of-the-art in security techniques, architectures, and algorithms for AI/ML in wireless communications.

Topics of interest

Topics of interest include, but are not limited to:

Adversarial ML Techniques

  • Adversarial examples
  • Poisoning attacks
  • Trojan/backdoor attacks
  • Generative adversarial learning
  • Defense techniques

Privacy and Security Issues of ML Solutions

  • Differential privacy
  • Information theoretic privacy
  • Physical layer privacy
  • Membership inference attacks
  • Model inversion

Strengthening ML Solutions

  • Datasets
  • Data augmentation
  • Privacy-preserving learning
  • Secure learning
  • Federated learning
  • Certified defense
  • Uncertainty quantification
  • Information discovery
  • Cognitive radio
  • Hardware solutions
  • Embedded computing
  • Experiments and testbeds

Relevant ML Applications

  • 5G/IoT security
  • Network slicing
  • Network virtualization
  • Anonymity
  • Authentication
  • Covert communications
  • Device identification
  • Intrusion detection
  • Localization
  • RF fingerprinting
  • Smart jamming and spoofing
  • Security for mobile autonomous multi-agent platforms

Workshop chairs

Deniz Gunduz
Imperial College London
London, UK
Brian Jalaian
U.S. Army Research Laboratory
Adelphi, MD, USA
Marc Kurz
University of Applied Sciences Upper Austria
Hagenberg, Austria
Berhard Moser
Software Competence Center Hagenberg
Hagenberg, Austria
Yalin E. Sagduyu
Intelligent Automation Inc.
Rockville, MD, USA
Yi Shi
Virginia Tech
Blacksburg, VA, USA
George Stantchev
U.S. Naval Research Laboratory
Washington, DC, USA

Keynotes

Threat Models for AI-Enabled 5G Infrastructure

Abstract 5G envisions using artificial intelligence and machine learning to dynamically orchestrate network resources to support real-time use cases, including near-real-time orchestration of the radio access network (RAN) and non-real-time orchestration of broader virtualized, software-defined network resources. This talk explores different approaches for applying AI/ML to these orchestration functions, and examines specific threat models to inform how these AI/ML functions should be secured.

Charles Clancy
The MITRE Corporation, Virginia, USA

Charles Clancy is Vice President for Intelligence Programs at MITRE where he leads the organization’s technical strategy and priorities in support of the US intelligence community and federal law enforcement. Before joining MITRE in 2019, Clancy served as the Bradley Professor of Cybersecurity in the Department of Electrical and Computer Engineering at Virginia Tech, and executive director at the Hume Center for National Security and Technology. He has previously served as an editor for IEEE Transactions on Information Forensics and Security, and IEEE Transactions on Cognitive Communications and Networking. He served as TPC co-chair for IEEE DySPAN 2017 and general chair for IEEE CNS 2019.

Side and Covert Channels: the Dr. Jekyll and Mr Hyde of Modern Technologies

While Smartphone and IoT devices usage become more and more pervasive, people start also asking to which extent such devices can be maliciously exploited as “tracking devices”. The concern is not only related to an adversary taking physical or remote control of the device, but also to what a passive adversary without the above capabilities can observe from the device communications. Work in this latter direction aimed, for example, at inferring the apps a user has installed on his device, or identifying the presence of a specific user within a network.
In this talk, we discuss threats coming from contextual information and to which extent it is feasible, for example, to identify the specific actions that a user is doing on mobile apps, by eavesdropping their encrypted network traffic. We will also discuss the possibility of building covert and side channels leveraging timing, heat, energy consumption, and audio signals, to steal information from mobile devices, as well as inferring keypresses, password & PINs.

Mauro Conti
University of Padua, Padua, Italy

Mauro Conti is Full Professor at the University of Padua, Italy, and also affiliated with TU Delft and UW Seattle. He obtained his Ph.D. from Sapienza University of Rome, Italy, in 2009. After his Ph.D., he was a Post-Doc Researcher at Vrije Universiteit Amsterdam, The Netherlands. In 2011 he joined as Assistant Professor the University of Padua, where he became Associate Professor in 2015, and Full Professor in 2018. He has been Visiting Researcher at GMU, UCLA, UCI, TU Darmstadt, UF, and FIU. He has been awarded with a Marie Curie Fellowship (2012) by the European Commission, and with a Fellowship by the German DAAD (2013). His research is also funded by companies, including Cisco, Intel, and Huawei. His main research interest is in the area of Security and Privacy. In this area, he published more than 300 papers in topmost international peer-reviewed journals and conference. He is Area Editor-in-Chief for IEEE Communications Surveys & Tutorials, and Associate Editor for several journals, including IEEE Communications Surveys & Tutorials, IEEE Transactions on Information Forensics and Security, IEEE Transactions on Dependable and Secure Computing, and IEEE Transactions on Network and Service Management. He was Program Chair for TRUST 2015, ICISS 2016, WiSec 2017, ACNS 2020, and General Chair for SecureComm 2012 and ACM SACMAT 2013. He is Senior Member of the IEEE.

Steering committee

  • Wenjing Lou, Virginia Tech, USA
  • Alan Michaels, Virginia Tech, USA
  • Stephen Russell, U.S. Army Research Laboratory, USA
  • K.P. (Suba) Subbalakshmi, Stevens Institute of Technology, USA
  • Sennur Ulukus, University of Maryland, USA

Submission guidelines

All papers for WiseML 2020 must be submitted electronically through the HotCRP conference management system.

Workshop extended abstracts must be written in English and are not to exceed three pages. Workshop papers must be written in English and are not to exceed six pages. Only PDF files will be accepted for the review process of both abstracts and papers. All submissions must be prepared using the following ACM proceedings template and must use US Letter page size (215.9 x 279.4 mm / 8.5 x 11 inches):

Accepted papers will appear in the conference proceedings and the ACM digital library.

Registration requirement

All accepted papers must be presented at the workshop. Therefore, at least one author of each accepted paper must register to the workshop at author rate (one per paper) and present the paper.

Important Dates

  • Extended abstract submission deadline: May 10, 2020 (23:59 AoE)
  • Extended abstract submission deadline (EXTENDED): May 12, 2020 (23:59 AoE)
  • Acceptance notification: May 18, 2020
  • Camera-ready paper submission: May 24, 2020 (23:59 AoE)
  • WiseML workshop: July 13, 2020