Deep Reconstruction-Classification Networks for Unsupervised Domain Adaptation about papers

[leo-p]

Summary:

• Dataset: SVHN, MNIST, USPS, CIFAR and STL.
• Objective: Build a network easily trainable by back-propagation to perform unsupervised domain adaptation while at the same time learning a good embedding for both source and target domains.

Architecture:

Very similar to RevGrad but with some differences.

Basically a shared encoder and then a classifier and a reconstructor.

The two losses are:

• the usual cross-entropy with softmax for the classifier
• the pixel-wise squared loss for reconstruction

Which are then combined using a trade-off hyper-parameter between classification and reconstruction.

They also use data augmentation to generate additional training data during the supervised training using only geometrical deformation: translation, rotation, skewing, and scaling

Plus denoising to reconstruct clean inputs given their noisy counterparts (zero-masked noise and Gaussian noise).

Results:

Outperforms state of the art on most tasks at the time, now outperformed itself by Generate To Adapt on most tasks.