Application of Neural Translation: Source Code to English
Abstract not provided.
Publications
Quantifying Uncertainty in Emulations: LDRD Report
This report summarizes the work performed under the project "Quantifying Uncertainty in Emulations." Emulation can be used to model real-world systems, typically using virtual- ization to run the real software on virtualized hardware. Emulations are increasingly used to answer mission-oriented questions, but how well they represent the real-world systems is still an open area of research. The goal of the project was to quantify where and how emulations differ from the real world. To do so, we ran a representative workload on both, and collected and compared metrics to identify differences. We aimed to capture behaviors, rather than performance, differences as the latter is more well-understood in the literature. This report summarizes the project's major accomplishments, with the background to understand these accomplishments. It gathers the abstracts and references for the refereed publications that have appeared as part of this work. We then archive partial work not yet ready for publication. 1 Principal Investigator 2 Remaining authors ordered alphabetically by last name
Using Sequence-to-Sequence Models to Build Source-to-English Translations
Abstract not provided.
Lessons Learned from 10k Experiments to Compare Virtual and Physical Testbeds
Abstract not provided.
Transformers for Neural Machine Translation and Beyond
Abstract not provided.
Exploiting Time and Subject Locality for Fast, Efficient, and Understandable Alert Triage
2018 International Conference on Computing, Networking and Communications, ICNC 2018
In many organizations, intrusion detection and other related systems are tuned to generate security alerts, which are then manually inspected by cyber-security analysts. These analysts often devote a large portion of time to inspecting these alerts, most of which are innocuous. Thus, it would be greatly beneficial to reduce the number of innocuous alerts, allowing analysts to utilize their time and skills for other aspects of cyber defense. In this work, we devise several simple, fast, and easily understood models to cut back this manual inspection workload, while maintaining high true positive and true negative rates. We demonstrate their effectiveness on real data, and discuss their potential utility in application by others.
Exploiting Time and Subject Locality for Fast Efficient and Understandable Alert Triage
Abstract not provided.
Exploiting locality for fast alert triage
Abstract not provided.
8 Results