CMU SEI Research Review 2019

CMU SEI Research Review 2019 Project Descriptions and Posters

November 1, 2019 • Annual Review

By Software Engineering Institute

This brochure includes descriptive information about the SEI's fiscal year 2019 research portfolio

Read

Research Review 2019 Schedule

October 21, 2019 • Brochure

By Software Engineering Institute

This handout outlines the schedule for Research Review 2019, held on October 21 and 22, 2019.

Learn More

Automated Code Repair to Ensure Memory Safety (2019)

October 28, 2019 • Presentation

By William Klieber

This presentation describes an approach to automatically repair source doe to assure memory safety.

Learn More

Automated Code Repair to Ensure Memory Safety in C Source Code (2019)

October 28, 2019 • Presentation

By William Klieber

This is a poster reflecting research to automatically repair C source code to eliminate memory-safety vulnerabilities.

Learn More

Automated Code Repair to Ensure Memory Safety (video)

November 11, 2019 • Video

By William Klieber

Watch Dr. Will Klieber discuss research on automatically repairing memory safety violations so programs become provably memory-safe.

Watch

Recovering Meaningful Variable Names in Decompiled Code

November 11, 2019 • Presentation

By Professor Vasilescu

In this project, we showed that the conventional wisdom that variable names cannot be recovered is wrong.

Learn More

Recovering Meaningful Variable Names in Decompiled Code (video)

November 11, 2019 • Video

By Professor Vasilescu

Watch CMU ISR assistant professor Dr. Bogdan Vasilescu discuss collaborative research with SEI in the reverse engineering of legacy systems using ML to identify variables.

Watch

Rapid Construction of Accurate Automatic Alert Handling System (2019)

October 28, 2019 • Presentation

By Lori Flynn, Ebonie McNeil

This poster includes information about research to develop a prototype source code analysis integrated framework environment.

Learn More

Rapid Construction of Accurate Automatic Alert Handling System (2019)

October 28, 2019 • Presentation

By Lori Flynn, Ebonie McNeil

This presentation describes development of a reference architecture and prototype for rapid, accurate, and adaptive alert classification and prioritization.

Learn More

Rapid Construction of Accurate Automatic Alert Handling System (video)

November 11, 2019 • Video

By Lori Flynn, Ebonie McNeil

Watch Dr. Lori Flynn and Ebonie McNeil discuss a prototype tool for automated static analysis alert classification that reduces cost and data barriers.

Watch

Causal Models for Software Cost Prediction and Control (video)

November 11, 2019 • Video

By Michael D. Konrad, Bill Nichols, Robert W. Stoddard, David Zubrow

Researchers at CMU SEI collaborated with other researchers to apply causal learning to learn how to control costs in software development and sustainment.

Watch

Causal Models for Software Cost Prediction and Control

October 28, 2019 • Presentation

By Michael D. Konrad, Robert W. Stoddard, Bill Nichols, David Zubrow

This presentation describes work toward establishing causation with observational data, which remains a vital need and a key technical challenge.

Learn More

Integrated Safety and Security Engineering for Mission-Critical Systems

October 28, 2019 • Presentation

By Sam Procter

This presentation describes research to develop software and processes that use a system's architecture to support developer intuition and improve safety and security.

Learn More

Graph Convolutional Neural Networks

October 28, 2019 • Presentation

By Oren Wright

This project used graph signal processing formalisms to create new deep learning tools for graph convolutional neural networks (GCNNs).

Learn More

Graph Convolutional Neural Networks

October 28, 2019 • Presentation

By Oren Wright

This poster describes research to use graph convolutional neural networks to extend deep learning.

Learn More

Graph Convolutional Neural Networks

November 12, 2019 • Video

By Oren Wright

Oren Wright discusses using graph signal processing formalisms to create new deep learning tools for graph convolutional neural networks (GCNNs).

Watch

KalKi: High Assurance Software-Defined IoT Security

October 28, 2019 • Presentation

By Sebastián Echeverría

This poster provides an update on research to help accelerate the safe use of Internet of Things devices in tactical settings.

Learn More

KalKi: High-Assurance Software-Defined IoT Security

October 28, 2019 • Presentation

By Sebastián Echeverría

We are developing a solution for DoD's use of commodity IoT technologies that remains resilient and trustworthy, even in the presence of a powerful attacker.

Learn More

Kalki: High Assurance Software-Defined IoT Security (video)

November 11, 2019 • Video

By Sebastián Echeverría

Watch Sebastian Echeverria discuss research to enable secure integration of IoT devices into DoD systems, even if devices are not fully trusted or configurable.

Watch

Video Summarization and Search

October 28, 2019 • Presentation

By Rachel Brower-Sinning, Jeffrey Hansen

This project's goal is to assist video analysts by increasing the volume of data that can be analyzed and providing insights into patterns of life

Learn More

Summarizing and Searching Video: Domain Adaptation

October 28, 2019 • Presentation

By Edwin J. Morris

This poster describes research to enable the use of existing data sets to training detectors in low data environments.

Learn More

Summarizing and Searching Video: Patterns-of-Life Analysis

October 28, 2019 • Presentation

By Edwin J. Morris

This poster describes experiments to use simulations to explore patterns-of-life analysis.

Learn More

Summarizing and Searching Video: Geometry-Aware Visual Surveillance

October 28, 2019 • Presentation

By Edwin J. Morris

This poster describes research to develop a moving-object tracker for video search.

Learn More

Projecting Quantum Computational Advantage Versus Classical State of the Art

October 28, 2019 • Presentation

By Jason Larkin

Learn More

Projecting Quantum Computational Advantage Versus Classical State of the Art

October 28, 2019 • Presentation

By Jason Larkin

This poster describes research to project when quantum computational advantage will outperform classical state of the art.

Learn More

Projecting Quantum Computational Advantage Versus Classical State of the Art (video)

November 11, 2019 • Video

By Jason Larkin

Watch

Untangling the Knot: Recommending Component Refactorings

October 28, 2019 • Presentation

By James Ivers

This presentation describes work to improve the ability to evolve software efficiently.

Learn More

Field Stripping a Weapons System: Building a Trustworthy Computer

October 28, 2019 • Presentation

By Gabriel Somlo

This poster describes research to determine the feasibility of rebuilding a weapons system from source code.

Learn More

Field Stripping a Weapons System: Building a Trustworthy Computer

October 28, 2019 • Presentation

By Gabriel Somlo

This presentation describes research to demonstrate the practicality of a comprehensive approach to guaranteeing the trustworthiness of computer systems.

Learn More

Field Stripping a Weapons System: Building a Trustworthy Computer (video)

November 11, 2019 • Video

By Gabriel Somlo

Watch SEI principal investigator Dr. Gabriel Somlo discuss research to demonstrate the practicality of guaranteeing the trustworthiness of DoD fielded computer systems.

Watch

Using All Processor Cores While Being Confident about Timing

October 28, 2019 • Presentation

By Bjorn Andersson

This presentation describes research to address the use of all processor cores while retaining confidence about timing.

Learn More

Using All Processor Cores While Being Confident about Timing

October 28, 2019 • Presentation

By Bjorn Andersson

This poster describes research on verifying software timing on multicore processors without full knowledge of memory system resources.

Learn More

Using All Processor Cores While Being Confident about Timing

November 11, 2019 • Video

By Bjorn Andersson

Watch Dr. Bjorn Andersson discuss research on solutions to avoid the common practice of disabling all processor cores except one in order to be confident about timing.

Watch

Emotion Recognition from Voice in the Wild

October 28, 2019 • Presentation

By Oren Wright

This poster represents research to determine the feasibility of using micro-articulometry technique to accurately recognize emotion from voice.

Learn More

Emotion Recognition from Voice in the Wild

October 28, 2019 • Presentation

By Oren Wright

This presentation describes progress in research to use advances in micro-articulometry to identify emotion from voice.

Learn More

Untangling the Knot: Recommending Component Refactorings

October 28, 2019 • Presentation

By James Ivers

This poster describes early research intended to outperform refactoring recommendations based only on quality metrics.

Learn More

Untangling the Knot: Recommending Component Refactorings

November 11, 2019 • Video

By James Ivers

Watch principal investigator James Ivers discuss research on applying AI techniques to rapidly increase the pace of making structural changes to software code bases.

Watch

A Series of Unlikely Events: Learning Patterns by Observing Sequential Behavior

October 28, 2019 • Presentation

By Eric Heim

This poster represents research to apply Inverse Reinforcement Learning techniques to model sequential behavior.

Learn More

A Series of Unlikely Events: Learning from Sequential Behavior for Activity-Based Intelligence and Modeling Human Expertise

October 28, 2019 • Presentation

By Eric Heim

This presentation describes work to use inverse reinforcement learning techniques to perform activity-based intelligence.

Learn More

A Series of Unlikely Events: Learning Patterns by Observing Sequential Behavior (video)

November 11, 2019 • Video

By Eric Heim

Watch SEI principal investigator Eric Heim discuss research to develop novel Inverse Reinforcement Learning (IRL) techniques.

Watch

Spiral/AIML: Co-optimization for High-Performance, Data-Intensive Computing in Resource Constrained Environments

October 28, 2019 • Presentation

By Scott McMillan, Franz Franchetti (Carnegie Mellon University)

This presentation updates research that helps developers build high-performance AI/ML applications on advanced hardware architectures faster.

Learn More

Spiral/AIML: Frontiers of Graph Processing in Linear Algebra

October 28, 2019 • Presentation

By Scott McMillan, Franz Franchetti (Carnegie Mellon University)

This poster describes research to use a linear algebraic approach to graph algorithms

Learn More

Spiral/AIML: Resource-Constrained Co-Optimization for High-Performance, Data-Intensive Computing

November 11, 2019 • Video

By Scott McMillan, Franz Franchetti (Carnegie Mellon University)

Watch SEI Principal Investigator, Dr. Scott McMillan, and research collaborator, CMU ECE Professor Franz Franchetti, discuss a community research effort to develop tools to reduce the prohibitive cost of implementing …

Watch

Rapid Certifiable Trust (video)

November 11, 2019 • Video

By Dionisio de Niz

Dr. Dionisio de Niz discusses verification techniques that reduce development and deployment time for cyber-physical systems, including unmanned vehicles.

Watch

Rapid Certifiable Trust

October 28, 2019 • Presentation

By Dionisio de Niz

Rapid Certifiable Trust preserves cyber-physical system safety by verifying and protecting part of a system and enforcing constraints on unverified components.

Learn More

Rapid Certifiable Trust

October 28, 2019 • Presentation

By Dionisio de Niz

This poster describes investigation into formal methods to automatically validate new technologies for safety.

Learn More

Panel: Supercharging the Software Factory: Deploying Capability at the Speed of Relevance

November 11, 2019 • Video

By Anita Carleton

Watch as Anita Carleton moderates a discussion on deploying mission capability at the speed of relevance to maintain DoD competitive advantage.

Watch

Integrating Safety and Security Engineering for Mission-Critical Systems

November 22, 2019 • Video

By Sam Procter, Alex Boydston (U.S. Army ADD/JMR)

Sam Procter discusses ongoing research to make it easier to specify, design, and assure critical systems that are safer and more secure using model based engineering with AADL.

Watch

Future Reach Conversation: Countering Adversarial Operations Made Possible by AI

November 22, 2019 • Video

By Allen D. Householder, Lujo Bauer (Carnegie Mellon University, Department of Electrical and Computer Engineering), Kathleen Carley (Carnegie Mellon School of Computer Science)

Watch as Dr. Matt Gaston, Director of SEI Emerging Technology Center, moderates discussion on countering adversarial operations made possible by AI

Watch

Summarizing and Searching Video (2019)

November 22, 2019 • Video

By Edwin J. Morris, Rachel Brower-Sinning, Jeffrey Hansen

SEI and CMU researchers discuss ways to improve the training of machine learning algorithms necessary for detecting, tracking, and recognizing patterns of objects in video.

Watch

Defining AI Engineering

February 14, 2020 • Video

By Matt Gaston

Dr. Matt Gaston and Professor Martial Hebert discuss the emerging field of AI Engineering.

Watch

Emotion Recognition from Voice in the Wild (Video)

February 14, 2020 • Video

By Oren Wright

Watch SEI Principal Investigator, Oren Wright, and CMU Language Technologies Institute collaborator, Shahan Ali Memon, discuss machine emotional intelligence and introduce an innovative speech emotion recognition database, CMU-SER.

Watch

Projecting How AI/ML Will Revolutionize Software Vulnerability Discovery

February 14, 2020 • Video

By Greg Shannon

Watch as Dr. Greg Shannon moderates discussion on projecting how AI/ML will revolutionize software vulnerability discovery, mitigation, and/or coordination.

Watch