Instead of increasing the fidelity of models for policy learning, we handle the distortion issue via learning to adapt to diverse simulators generated by the offline dataset.
At last, we proposed a differentiable auto data augmentation method to further improve estimation accuracy.
DEMER also derives a recommendation policy with a significantly improved performance in the test phase of the real application.
We assume that problems inside are inadequate use of supervision information and imbalance between semantics and details at all level feature maps in CNN even with Feature Pyramid Networks (FPN).
In this work, we introduce the notion of image retargetability to describe how well a particular image can be handled by content-aware image retargeting.