Deep Supervision with Intermediate Concepts (BayLearn)
Read Deep Supervision with Intermediate Concepts (BayLearn). We introduce a novel technique for training convolutional neural networks (CNNs), namely deep supervision with intermediate concepts, leading to improved generalization. Our approach draws inspiration from Deeply Supervised Nets (DSN) [5], which supervises each layer by the main task to accelerate training convergence. Our method differs from DSN in that we apply deep supervision with intermediate concepts, intrinsic to the ultimate task, to regularize the network for better generalization. We apply this improved generalization ability to transfer knowledge from synthetic to real images.
