CFP: ICML 2019 Workshop on Uncertainty and Robustness in Deep Learning
Balaji Lakshminarayanan
ICML 2019 Workshop: Uncertainty and Robustness in Deep Learning
Long Beach, California, USA
June 14 or 15, 2019
https://sites.google.com/view/udlworkshop2019/
DESCRIPTION:
There has been growing interest in making deep neural networks robust for real-world applications. Challenges arise when models receive inputs drawn from outside the training distribution. For example, a neural network tasked with classifying handwritten digits may assign high confidence predictions to cat images. Anomalies are frequently encountered when deploying ML models in the real world. Generalization to unseen and worst-case inputs is also essential for robustness to distributional shift. Well-calibrated predictive uncertainty estimates are indispensable for many machine learning applications, such as self-driving cars and medical diagnosis systems. In order to have ML models reliably predict in open environment, we must deepen technical understanding in the following areas: (1) learning algorithms that are robust to changes in input data distribution (e.g., detect out-of-distribution examples); (2) mechanisms to estimate and calibrate confidence produced by neural networks and (3) methods to improve robustness to adversarial and non-adversarial corruptions, and (4) key applications for uncertainty (e.g., computer vision, robotics, self-driving cars, medical imaging) as well as broader machine learning tasks.
We invite the submission of papers on topics including, but not limited to:
Out-of-distribution detection and anomaly detection
Robustness to corruptions, adversarial perturbations, and distribution shift
Calibration
Probabilistic (Bayesian and non-Bayesian) neural networks
Open world recognition and open set learning
Security
Quantifying different types of uncertainty (known unknowns and unknown unknowns) and types of robustness
Applications of robust and uncertainty-aware deep learning
KEY DATES:
Submission deadline: April 30, 2019*
Author notification: May 15, 2019*
Camera ready deadline: May 30, 2019
Workshop date: June 14 or 15, 2019 (TBA)
*If you need a decision sooner (e.g. to apply for visa and/or plan your trip), please submit earlier and we can expedite the review upon request.
SUBMISSION INSTRUCTIONS:
Researchers interested in contributing should upload a short paper of up to 4 pages in PDF format by April 30, 2019, 11:59pm (anywhere on earth) using the form at https://sites.google.com/view/udlworkshop2019/call-for-papers. References and supplementary material can exceed 4 pages. Authors should use the style files in this zip file. Submissions don't need to be anonymized.
If you have any questions, please contact us at udlwork...@gmail.com.
ORGANIZERS:
Yixuan (Sharon) Li (Facebook AI)
Balaji Lakshminarayanan (Google DeepMind)
Dan Hendrycks (UC Berkeley)
Thomas Dietterich (Oregon State University)
Justin Gilmer (Google Brain)