Transfer Learning in CNNs Using Filter-Trees
This addresses efficiency in CNN training for pattern recognition tasks, but it is incremental as it builds on existing transfer learning methods.
The paper tackles the problem of training deep CNNs requiring extensive computation and large data by proposing Bank of Filter-Trees (BFT), a transfer learning mechanism that transfers subnetworks from multiple source networks to a new task, achieving performance on par with networks trained from scratch without finetuning.
Convolutional Neural Networks (CNNs) are very effective for many pattern recognition tasks. However, training deep CNNs needs extensive computation and large training data. In this paper we propose Bank of Filter-Trees (BFT) as a trans- fer learning mechanism for improving efficiency of learning CNNs. A filter-tree corresponding to a filter in k^{th} convolu- tional layer of a CNN is a subnetwork consisting of the filter along with all its connections to filters in all preceding layers. An ensemble of such filter-trees created from the k^{th} layers of many CNNs learnt on different but related tasks, forms the BFT. To learn a new CNN, we sample from the BFT to select a set of filter trees. This fixes the target net up to the k th layer and only the remaining network would be learnt using train- ing data of new task. Through simulations we demonstrate the effectiveness of this idea of BFT. This method constitutes a novel transfer learning technique where transfer is at a sub- network level; transfer can be effected from multiple source networks; and, with no finetuning of the transferred weights, the performance achieved is on par with networks that are trained from scratch.