235 citations · 850 across the 47 of their papers we have counts for
14 papers · 1 filter
A Closer Look at Robustness to L-infinity and Spatial Perturbations and their Composition
Luke Rowe, Benjamin Thérien, Krzysztof Czarnecki +1
In adversarial machine learning, the popular threat model has been the focus of much previous work. While this mathematical definition of imperceptibility successfull…
A Safety Assurable Human-Inspired Perception Architecture
Rick Salay, Krzysztof Czarnecki
Although artificial intelligence-based perception (AIP) using deep neural networks (DNN) has achieved near human level performance, its well-known limitations are obstacles to the…
Recursive Constraints to Prevent Instability in Constrained Reinforcement Learning
Jaeyoung Lee, Sean Sedwards, Krzysztof Czarnecki
We consider the challenge of finding a deterministic policy for a Markov decision process that uniformly (in all states) maximizes one reward subject to a probabilistic constraint…
The missing link: Developing a safety case for perception components in automated driving
Rick Salay, Krzysztof Czarnecki, Hiroshi Kuwajima +6
Safety assurance is a central concern for the development and societal acceptance of automated driving (AD) systems. Perception is a key aspect of AD that relies heavily on Machine…
The Effect of Optimization Methods on the Robustness of Out-of-Distribution Detection Approaches
Vahdat Abdelzad, Krzysztof Czarnecki, Rick Salay
Deep neural networks (DNNs) have become the de facto learning mechanism in different domains. Their tendency to perform unreliably on out-of-distribution (OOD) inputs hinders their…
Detecting Out-of-Distribution Inputs in Deep Neural Networks Using an Early-Layer Output
Vahdat Abdelzad, Krzysztof Czarnecki, Rick Salay +3
Deep neural networks achieve superior performance in challenging tasks such as image classification. However, deep classifiers tend to incorrectly classify out-of-distribution (OOD…
Out-of-distribution Detection in Classifiers via Generation
Sachin Vernekar, Ashish Gaurav, Vahdat Abdelzad +3
By design, discriminatively trained neural network classifiers produce reliable predictions only for in-distribution samples. For their real-world deployments, detecting out-of-dis…
A Micro-Objective Perspective of Reinforcement Learning
Changjian Li, Krzysztof Czarnecki
The standard reinforcement learning (RL) formulation considers the expectation of the (discounted) cumulative reward. This is limiting in applications where we are concerned with n…
Analysis of Confident-Classifiers for Out-of-distribution Detection
Sachin Vernekar, Ashish Gaurav, Taylor Denouden +4
Discriminatively trained neural classifiers can be trusted, only when the input data comes from the training distribution (in-distribution). Therefore, detecting out-of-distributio…
WiseMove: A Framework for Safe Deep Reinforcement Learning for Autonomous Driving
Jaeyoung Lee, Aravind Balakrishnan, Ashish Gaurav +2
Machine learning can provide efficient solutions to the complex problems encountered in autonomous driving, but ensuring their safety remains a challenge. A number of authors have…
Improving Reconstruction Autoencoder Out-of-distribution Detection with Mahalanobis Distance
Taylor Denouden, Rick Salay, Krzysztof Czarnecki +3
There is an increasingly apparent need for validating the classifications made by deep learning systems in safety-critical applications like autonomous vehicle systems. A number of…
Calibrating Uncertainties in Object Localization Task
Buu Phan, Rick Salay, Krzysztof Czarnecki +3
In many safety-critical applications such as autonomous driving and surgical robots, it is desirable to obtain prediction uncertainties from object detection modules to help suppor…
Urban Driving with Multi-Objective Deep Reinforcement Learning
Changjian Li, Krzysztof Czarnecki
Autonomous driving is a challenging domain that entails multiple aspects: a vehicle should be able to drive to its destination as fast as possible while avoiding collision, obeying…
Using Machine Learning Safely in Automotive Software: An Assessment and Adaption of Software Process Requirements in ISO 26262
Rick Salay, Krzysztof Czarnecki
The use of machine learning (ML) is on the rise in many sectors of software development, and automotive software development is no different. In particular, Advanced Driver Assista…