Competitive Learning with Feedforward Supervisory Signal for Pre-trained Multilayered Networks
This work addresses the challenge of improving robustness and representation in pre-trained networks, which is incremental as it builds on existing methods with a new supervisory approach.
The authors tackled the problem of robust learning and revising internal representations in pre-trained multilayered neural networks by proposing a novel method that uses feedforward supervisory signals and associates new inputs with pre-trained ones, effectively leveraging rich input information from earlier layers.
We propose a novel learning method for multilayered neural networks which uses feedforward supervisory signal and associates classification of a new input with that of pre-trained input. The proposed method effectively uses rich input information in the earlier layer for robust leaning and revising internal representation in a multilayer neural network.