Transfer learning using ImageNet pre-trained models has been the de facto approach in a wide range of computer vision tasks . However, fine-tuning still requires task-specific training data… In this paper, we propose a new paradigm of synthesizing neural networks from language descriptions . We also demonstrate a simple method to help identify keywords in language descriptions leveraged by \textbf{N3} when synthesizing model parameters . To the best of our knowledge, this is the first method to . synthesize entire neural networks . from natural . language descriptions to generate parameter adaptations as well as a new task . classification layer for a pre- trained neural network, effectively {“}fine-tuned the network for a new . task .

Links: PDF - Abstract

Code :

None

Keywords : language - descriptions - neural - task - network -

Leave a Reply

Your email address will not be published. Required fields are marked *