I am asking this question on deep neural network architectures only. If you want to restrict the domain of tasks then you can choose computer vision for this question.

Suppose there is an architecture that performs well on a task. Is it possible can edit or append the first or last few layers and then it performs similarly well on the other task?

If yes, please provide me an example of such architecture that performs well on at least a couple of tasks.

  • 1
    $\begingroup$ It seems that you're looking for something similar to transfer learning. Let me know if that's the case or not. $\endgroup$
    – nbro
    Jan 4 at 13:46
  • $\begingroup$ A pre-trained model is a saved network that was previously trained on a large dataset, typically on a large-scale image-classification task. You either use the pretrained model as is or use transfer learning to customize this model to a given task. #1/#2 $\endgroup$
    – hanugm
    Jan 4 at 22:13
  • $\begingroup$ Fine-Tuning: Unfreeze a few of the top layers of a frozen model base and jointly train both the newly-added classifier layers and the last layers of the base model. This allows us to "fine-tune" the higher-order feature representations in the base model in order to make them more relevant for the specific task. #2/#2 $\endgroup$
    – hanugm
    Jan 4 at 22:14
  • $\begingroup$ Yeah @nbro. It is the same-thing I am looking for. It seems that task is not fixed and hence we can use pre-trained model for any task under consideration. $\endgroup$
    – hanugm
    Jan 4 at 22:15

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.