Baseline experiments clarification

[ShaniGam]

Great work!
I'm looking at the experiments in the paper and I'm a bit confused. As far as I understand, your source consists of examples from the source distribution+(fewer) examples from the target distribution. If that's the case, does the ResNet-50 baseline also includes examples from the target distribution or only examples from the source distribution?

[tim-learn]

Hi, @ShaniGam

The ResNet-50 baseline is a source-only method that does not include examples from the target distribution.

Best

Tim

[ShaniGam]

Did you compare your work to a network that was trained on the same exact data (source+partial target)?

[tim-learn]

@ShaniGam Since the target data is unlabeled, it is hard to find a solution to use it during supervised training with source data. Our paper aims to include the target data into the source domain during alignment instead of classification.

[ShaniGam]

I thought you used some labeled data for training in your method but I guess I was wrong. thanks.