Adaptive Routing Between Capsules
This work addresses computational and scalability issues in capsule networks, which is an incremental improvement for deep learning researchers.
The paper tackles the limitations of dynamic routing in capsule networks, such as inability to stack multiple layers and high computational cost, by proposing an adaptive routing algorithm that achieves better performance on datasets like CIFAR10 and Fashion-MNIST.
Capsule network is the most recent exciting advancement in the deep learning field and represents positional information by stacking features into vectors. The dynamic routing algorithm is used in the capsule network, however, there are some disadvantages such as the inability to stack multiple layers and a large amount of computation. In this paper, we propose an adaptive routing algorithm that can solve the problems mentioned above. First, the low-layer capsules adaptively adjust their direction and length in the routing algorithm and removing the influence of the coupling coefficient on the gradient propagation, so that the network can work when stacked in multiple layers. Then, the iterative process of routing is simplified to reduce the amount of computation and we introduce the gradient coefficient $λ$. Further, we tested the performance of our proposed adaptive routing algorithm on CIFAR10, Fashion-MNIST, SVHN and MNIST, while achieving better results than the dynamic routing algorithm.