Natural target functions and tasks typically exhibit hierarchical modularity -- they can be broken down into simpler sub-functions that are organized in a hierarchy. Such sub-functions have two important features: they have a distinct set of inputs (input-separability) and they are reused as inputs higher in the hierarchy (reusability). Previous studies have established that hierarchically modular neural networks, which are inherently sparse, offer benefits such as learning efficiency, generalization, multi-task learning, and transfer. However, identifying the underlying sub-functions and their hierarchical structure for a given task can be challenging. The high-level question in this work is: if we learn a task using a sufficiently deep neural network, how can we uncover the underlying hierarchy of sub-functions in that task? As a starting point, we examine the domain of Boolean functions, where it is easier to determine whether a task is hierarchically modular. We propose an approach based on iterative unit and edge pruning (during training), combined with network analysis for module detection and hierarchy inference. Finally, we demonstrate that this method can uncover the hierarchical modularity of a wide range of Boolean functions and two vision tasks based on the MNIST digits dataset.

Methods that sparsify a network at initialization are important in practice because they greatly improve the efficiency of both learning and inference. Our work is based on a recently proposed decomposition of the Neural Tangent Kernel (NTK) that has decoupled the dynamics of the training process into a data-dependent component and an architecture-dependent kernel {–} the latter referred to as Path Kernel. That work has shown how to design sparse neural networks for faster convergence, without any training data, using the Synflow-L2 algorithm. We first show that even though Synflow-L2 is optimal in terms of convergence, for a given network density, it results in sub-networks with “bottleneck” (narrow) layers {–} leading to poor performance as compared to other data-agnostic methods that use the same number of parameters. Then we propose a new method to construct sparse networks, without any training data, referred to as Paths with Higher-Edge Weights (PHEW). PHEW is a probabilistic network formation method based on biased random walks that only depends on the initial weights. It has similar path kernel properties as Synflow-L2 but it generates much wider layers, resulting in better generalization and performance. PHEW achieves significant improvements over the data-independent SynFlow and SynFlow-L2 methods at a wide range of network densities.

The detection of the region of interests (ROIs) on Whole Slide Images (WSIs) is one of the primary steps in computer-aided cancer diagnosis and grading. Early and accurate identification of invasive cancer regions in WSI is critical in the improvement of breast cancer diagnosis and further improvements in patient survival rates. However, invasive cancer ROI segmentation is a challenging task on WSI because of the low contrast of invasive cancer cells and their high similarity in terms of appearance, to non-invasive regions. In this paper, we propose a CNN based architecture for generating ROIs through segmentation. The network tackles the constraints of data-driven learning and working with very low-resolution WSI data in the detection of invasive breast cancer. Our proposed approach is based on transfer learning and the use of dilated convolutions. We propose a highly modified version of U-Net based auto-encoder, which takes as input an entire WSI with a resolution of 320×320. The network was trained on low-resolution WSI from four different data cohorts and has been tested for inter as well as intra- dataset variance. The proposed architecture shows significant improvements in terms of accuracy for the detection of invasive breast cancer regions.

The aforementioned works and other analogous studies in finger knuckle images recognition have claimed that the precise detection of true features is difficult from poorly segmented images and the main reason for matching errors. Thus, an accurate segmentation of the region of interest is very crucial to achieve superior recognition results. In this paper, we have proposed a novel holistic and generalized segmentation Network (HFDSegNet) that automatically categorizes the given finger dorsal image obtained from multiple sensory resources into particular class and then extracts three possible ROIs (major knuckle, minor knuckle and nail) accurately. To best of our knowledge, this is the first attempt, an end-to-end trained object detector inspired by Deep Learning technique namely faster R-CNN (Region based Convolutional Neural Network) has been employed to detect and localize the position of finger knuckles and nail, even finger images exhibit blur, occlusion, low contrast etc. The experimental results are examined on two publicly available databases named as Poly-U contact-less FKI data-set, and Poly U FKP database. The proposed network is trained only over 500 randomly selected images per database, demonstrate the outstanding performance of proposed ROI’s segmentation network.