Mask selective regularization for restricted Boltzmann machines

Abstract

In the present work, we propose to deal with two important issues regarding to the RBM's learning capabilities. First, the topology of the input space, and second, the sparseness of the RBM obtained. One problem of RBMs is that they do not take advantage of the topology of the input space. In order to alleviate this lack, we propose to use a surrogate of the mutual information of the input representation space to build a set of binary masks. This approach is general and not only applicable to images, thus it can be extended to other layers in the standard layer-by-layer unsupervised learning. On the other hand, we propose a selective application of two different regularization terms, L 1 and L 2 , in order to ensure the sparseness of the representation and the generalization capabilities. Additionally, another interesting capability of our approach is the adaptation of the topology of the network during the learning phase by means of selecting the best set of binary masks that fit the current weights configuration.The performance of these new ideas is assessed with a set of experiments on different well-known corpus.

High Accuracy Thyroid Tumor Image Recognition Based on Hybrid Multiple Models Optimization

“…Such as Segmentation of Spine on MR Image by 3D CNN from Korez et al [18] and segmentation of brain and chest muscles on MRI images from Moeskops et al [19]. In classification and recognition field, there are detection of retinal base in diabetic patients by transfer learning from Gulshan et al [20], classification of skin cancer by transfer learning from Esteva et al [21], automatic classification on DCE-MRI by Sparse Autoencoder from Mansanet et al [22] and the detection and evaluation of the grade of knee osteoarthritis using CNN on X-ray image from Antony et al [23]. Reddy, G. T etc.…”