2
$\begingroup$

How does transfer learning work for regression tasks? Can someone point to an application where transfer learning has been successfully applied for regression tasks.

$\endgroup$
3
  • 1
    $\begingroup$ Do you need it to be regression to regression transfer learning or can be from classification to regression? The latter is quite common actually. $\endgroup$ Dec 7, 2019 at 10:35
  • $\begingroup$ It would be great to know about both. If you can shed some light on either/both or point me to an application/papers that would be great. $\endgroup$
    – GSH
    Dec 8, 2019 at 15:50
  • $\begingroup$ kaggle.com/c/diabetic-retinopathy-detection and kaggle.com/c/aptos2019-blindness-detection is one example of classification to regression. $\endgroup$ Dec 9, 2019 at 3:18

1 Answer 1

1
$\begingroup$

Consider the denoising autoencoder. The input features to a neural network are contaminated by a small amount of noise, sent through one or more intermediate layers, and then through a final layer of the same size as the input layer. This network is optimized to reconstruct the original data which can be seen as a form of regularization.

The resulting network weights (minus the final layer) are frozen and transferred to a supervised learning task (perhaps added to another neural network). It has been found that the features constructed by this process are quite valuable. This applies to regression or classification.

$\endgroup$
1
  • 1
    $\begingroup$ Could you add some references/links/examples to this answer? $\endgroup$ Aug 6, 2020 at 16:45

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.