Visible to the public KU SoS Lablet Quarterly Executive Summary - 2020 Q4Conflict Detection Enabled

A. Fundamental Research

The University of Kansas Lablet continued work on five projects on resiliency, IoT and cloud privacy, preventing side channel communication, and developing semantics and infrastructure for trust and initiated a fifth project on secure native binary execution. Specifically, we are: (i) developing a method to enable cloud-assisted, privacy-preserving machine learning classification over encrypted data for IoT devices; (ii) reducing micro-architectural side-channels by introducing new OS abstractions while minimally modifying micro-architecture and OS; (iii) developing an epistemology and ontology for framing resilience; (iv) formalizing the remote attestation and defining sufficiency and soundness; and (v) developing a framework for client-side security assessment and enforcement for COTS software. 

B. Community Engagement(s)

KU Lablet PIs participated in a 2-day cybersecurity workshop hosted at Maple Ranch by Enterprise Kansas City and Cliff Ellig, founder of Cerner, inc.  The objective of this workshop was to begin developing a long-range plan to enhance cybersecurity work in the Kansas City metropolitan area.  Workforce development and research were primary topics.  Enterprise KC estimates over 2300 unfilled cybersecurity positions regionally.  If those positions remain unfilled we will have difficulty keeping companies in the midwest.  We discussed what KU can do to increase training opportunities from certificate-level programs through PhD programs.

We presented our Lablet research projects and defined infrastructure required to grow our capabilities.  We discussed projects ranging from traditional funded research to industry funded research, technology transfer, and developing national centers of excellence.  Lablet research projects are catalyst for this effort and our presence at the workshop both defined where we are headed and enabled outreach to Kansas City businesses.

John Symons and his colleagues Ramón Alvarado (University of Oregon) and Kamuran Osmanoglu (Koç University) continued their international conference series on data privacy.  The workshop has attracted a broad international attention with attendance frequently exceeding 100 participants.  Presentations this quarter included:

  • “Data technology, policy and governance: what is the role of data science in democratic society and can we trust it?”, Thursday, October 22, 2020 at 11 a.m. (Pacific), Speaker: Denisa Kera (University of Salamanca)
  • “Data and us: forms, algorithms and the data subject”, Wednesday, November 18, 2020 at 11 a.m. (Pacific), Speaker: Colin Koopman (University of Oregon)
  • “Data Science in Scientific Inquiry, Philosophical issues”, Wednesday, December 9, 2020 at 11 a.m. (Pacific), Speaker: Sabina Leonelli (University of Exeter)

We continue planning to co-sponsor GenCyber and CyberBlitz workshops for primary and secondary school teachers, and active military members from Ft. Riley and high school students, respectively.  Details of the workshops are fluid due to public health issues, but we hope to hold both events in-person late this summer.

The KU Science of Security Advisory Board delayed its planned meeting last quarter, but is planning a February meeting and a meeting in associated with HoTSoS.  We have experience some turnover within the group with Aaron Wiessenfluh moving from Cboe to American Century.  We hope to add American Century to our board.

C. Educational Advances

As a result of our cybersecurity workshop with Enterprise Kansas City and Cliff Illig several Lablet researchers and leadership from the KU Achievement and Assessment Institute (AAI) began a process of building a knowledge map of cybersecurity topics.  AAI is a designated KU research center that is among the nation’s largest providers of state-wide testing services.  It is hoped this initial effort will result in AAI developing new, flexible mechanisms for cybersecurity education and certification.  This work is new and technically outside our Lablet research, but involves our PIs and shows significant promise.