Designing safety critical software systems to manage inherent uncertainty
AC Serban - 2019 IEEE International Conference on Software …, 2019 - ieeexplore.ieee.org
Deploying machine learning algorithms in safety critical systems raises new challenges for
system designers. The opaque nature of some algorithms together with the potentially large …
system designers. The opaque nature of some algorithms together with the potentially large …
[PDF][PDF] An architectural risk analysis of machine learning systems: Toward more secure machine learning
G McGraw, H Figueroa, V Shepardson… - Berryville Institute of …, 2020 - garymcgraw.com
At BIML, we are interested in “building security in” to machine learning (ML) systems from a
security engineering perspective. This means understanding how ML systems are designed …
security engineering perspective. This means understanding how ML systems are designed …
Multilayered review of safety approaches for machine learning-based systems in the days of AI
The unprecedented advancement of artificial intelligence (AI) in recent years has altered our
perspectives on software engineering and systems engineering as a whole. Nowadays …
perspectives on software engineering and systems engineering as a whole. Nowadays …
Unifying evaluation of machine learning safety monitors
With the increasing use of Machine Learning (ML) in critical autonomous systems, runtime
monitors have been developed to detect prediction errors and keep the system in a safe …
monitors have been developed to detect prediction errors and keep the system in a safe …
Engineering safety in machine learning
KR Varshney - 2016 Information Theory and Applications …, 2016 - ieeexplore.ieee.org
Machine learning algorithms are increasingly influencing our decisions and interacting with
us in all parts of our daily lives. Therefore, just like for power plants, highways, and myriad …
us in all parts of our daily lives. Therefore, just like for power plants, highways, and myriad …
Towards automated security design flaw detection
Efficiency of security-by-design has become an important goal for organizations
implementing software engineering practices such as Agile, DevOps, and Continuous …
implementing software engineering practices such as Agile, DevOps, and Continuous …
Smartchoices: Augmenting software with learned implementations
We are living in a golden age of machine learning. Powerful models perform many tasks far
better than is possible using traditional software engineering approaches alone. However …
better than is possible using traditional software engineering approaches alone. However …
Machine learning system architectural pattern for improving operational stability
H Yokoyama - 2019 IEEE International Conference on Software …, 2019 - ieeexplore.ieee.org
Recently, machine learning systems with inference engines have been widely used for a
variety of purposes (such as prediction and classification) in our society. While it is quite …
variety of purposes (such as prediction and classification) in our society. While it is quite …
Enforcing architectural security decisions
S Jasser - 2020 IEEE International Conference on Software …, 2020 - ieeexplore.ieee.org
Software architects should specify security measures for a software system on an
architectural level. However, the implementation often diverges from this intended …
architectural level. However, the implementation often diverges from this intended …
Robust machine learning systems: Reliability and security for deep neural networks
Machine learning is commonly being used in almost all the areas that involve advanced
data analytics and intelligent control. From applications like Natural Language Processing …
data analytics and intelligent control. From applications like Natural Language Processing …