2 code implementations • 23 Oct 2019 • Vahdat Abdelzad, Krzysztof Czarnecki, Rick Salay, Taylor Denounden, Sachin Vernekar, Buu Phan
Several approaches have been proposed to detect OOD inputs, but the detection task is still an ongoing challenge.
1 code implementation • 9 Oct 2019 • Sachin Vernekar, Ashish Gaurav, Vahdat Abdelzad, Taylor Denouden, Rick Salay, Krzysztof Czarnecki
By design, discriminatively trained neural network classifiers produce reliable predictions only for in-distribution samples.
Out-of-Distribution Detection Out of Distribution (OOD) Detection
1 code implementation • 25 Sep 2019 • Sachin Vernekar, Ashish Gaurav, Vahdat Abdelzad, Taylor Denouden, Rick Salay, Krzysztof Czarnecki
In the context of OOD detection for image classification, one of the recent approaches proposes training a classifier called “confident-classifier” by minimizing the standard cross-entropy loss on in-distribution samples and minimizing the KLdivergence between the predictive distribution of OOD samples in the low-density“boundary” of in-distribution and the uniform distribution (maximizing the entropy of the outputs).
Out-of-Distribution Detection Out of Distribution (OOD) Detection
1 code implementation • 27 Apr 2019 • Sachin Vernekar, Ashish Gaurav, Taylor Denouden, Buu Phan, Vahdat Abdelzad, Rick Salay, Krzysztof Czarnecki
Discriminatively trained neural classifiers can be trusted, only when the input data comes from the training distribution (in-distribution).
no code implementations • 6 Dec 2018 • Taylor Denouden, Rick Salay, Krzysztof Czarnecki, Vahdat Abdelzad, Buu Phan, Sachin Vernekar
There is an increasingly apparent need for validating the classifications made by deep learning systems in safety-critical applications like autonomous vehicle systems.
no code implementations • 27 Nov 2018 • Buu Phan, Rick Salay, Krzysztof Czarnecki, Vahdat Abdelzad, Taylor Denouden, Sachin Vernekar
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 support safe decision-making.