`

Timezone: »

 
VisDA21: Visual Domain Adaptation + Q&A
Kate Saenko · Kuniaki Saito · Donghyun Kim · Samarth Mishra · Ben Usman · Piotr Teterwak · Dina Bashkirova · Dan Hendrycks

Tue Dec 07 10:45 AM -- 11:05 AM (PST) @ None
in Competition Track Day 1: Overviews + Breakout Sessions »
Event URL: http://ai.bu.edu/visda-2021/ »

Progress in machine learning is typically measured by training and testing a model on the same distribution of data, i.e., the same domain. However, in real world applications, models often encounter out-of-distribution data. The VisDA21 competition invites methods that can adapt to novel test distributions and handle distributional shifts. Our task is object classification, but we measure accuracy on novel domains, rather than the traditional in-domain benchmarking. Teams will be given labeled source data and unlabeled target data from a different distribution (such as novel viewpoints, backgrounds, image quality). In addition, the target data may have missing and/or novel classes. Successful approaches will improve classification accuracy of known categories on target-domain data while learning to deal with missing and/or unknown categories.

Author Information

Kate Saenko (Boston University & MIT-IBM Watson AI Lab, IBM Research)
Kuniaki Saito (Boston University)
Donghyun Kim (Boston University)
Samarth Mishra (Boston University)
Ben Usman (Boston University)
Piotr Teterwak (Boston University)
Dina Bashkirova (Boston University)
Dan Hendrycks (UC Berkeley)

More from the Same Authors