In my experience, pre-trained models are not suitable for unsupervised tasks. Especially in deep clustering when pre-trained models are used, they often have worse results than without pre-trained models.
What is the scientific reason for this?
Why are the learned representations in pre-trained models not suitable for clustering?
Dear Amin Golzari Oskouei ,
Look the link, maybe useful.
https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=8412085
Regards,
Shafagat
If you are processing your data for your new research, I suggest, never use pre-trained model. For a new research process you need to pre-processes your data according to your models/methods/algorithms as a best match. Accordingly, you will have satisfactory and fair results. In fact no pretrained models are universal by any way.
- Badges
- Science topic
- Similar topics
- Psychology
- Mental Processes
- Learning