Branch Convolutional Neural Network (B-CNN)
Convolutional Neural Network (CNN) image classifiers are traditionally designed to have sequential convolutional layers with a single output layer. This is based on the assumption that all target classes should be treated equally and exclusively. However, some classes can be more difficult to distinguish than others, and classes may be organized in a hierarchy of categories. At the same time, a CNN is designed to learn internal representations that abstract from the input data based on its hierarchical layered structure. So it is natural to ask if an inverse of this idea can be applied to learn a model that can predict over a classification hierarchy using multiple output layers in decreasing order of class abstraction. In this paper, we introduce a variant of the traditional CNN model named the Branch Convolutional Neural Network (B-CNN). A B-CNN model outputs multiple predictions ordered from coarse to fine along the concatenated convolutional layers corresponding to the hierarchical structure of the target classes, which can be regarded as a form of prior knowledge on the output. To learn with B-CNNs a novel training strategy, named the Branch Training strategy (BT-strategy), is introduced which balances the strictness of the prior with the freedom to adjust parameters on the output layers to minimize the loss. In this way we show that CNN based models can be forced to learn successively coarse to fine concepts in the internal layers at the output stage, and that hierarchical prior knowledge can be adopted to boost CNN models’ classification performance. Our models are evaluated to show that the B-CNN extensions improve over the corresponding baseline CNN on the benchmark datasets MNIST, CIFAR-10 and CIFAR-100. …

Autonomous Deep Learning (ADL)
The feasibility of deep neural networks (DNNs) to address data stream problems still requires intensive study because of the static and offline nature of conventional deep learning approaches. A deep continual learning algorithm, namely autonomous deep learning (ADL), is proposed in this paper. Unlike traditional deep learning methods, ADL features a flexible structure where its network structure can be constructed from scratch with the absence of initial network structure via the self-constructing network structure. ADL specifically addresses catastrophic forgetting by having a different-depth structure which is capable of achieving a trade-off between plasticity and stability. Network significance (NS) formula is proposed to drive the hidden nodes growing and pruning mechanism. Drift detection scenario (DDS) is put forward to signal distributional changes in data streams which induce the creation of a new hidden layer. Maximum information compression index (MICI) method plays an important role as a complexity reduction module eliminating redundant layers. The efficacy of ADL is numerically validated under the prequential test-then-train procedure in lifelong environments using nine popular data stream problems. The numerical results demonstrate that ADL consistently outperforms recent continual learning methods while characterizing the automatic construction of network structures. …

Graph-Based Global Reasoning Network
Globally modeling and reasoning over relations between regions can be beneficial for many computer vision tasks on both images and videos. Convolutional Neural Networks (CNNs) excel at modeling local relations by convolution operations, but they are typically inefficient at capturing global relations between distant regions and require stacking multiple convolution layers. In this work, we propose a new approach for reasoning globally in which a set of features are globally aggregated over the coordinate space and then projected to an interaction space where relational reasoning can be efficiently computed. After reasoning, relation-aware features are distributed back to the original coordinate space for down-stream tasks. We further present a highly efficient instantiation of the proposed approach and introduce the Global Reasoning unit (GloRe unit) that implements the coordinate-interaction space mapping by weighted global pooling and weighted broadcasting, and the relation reasoning via graph convolution on a small graph in interaction space. The proposed GloRe unit is lightweight, end-to-end trainable and can be easily plugged into existing CNNs for a wide range of tasks. Extensive experiments show our GloRe unit can consistently boost the performance of state-of-the-art backbone architectures, including ResNet, ResNeXt, SE-Net and DPN, for both 2D and 3D CNNs, on image classification, semantic segmentation and video action recognition task. …

FanStore
Emerging Deep Learning (DL) applications introduce heavy I/O workloads on computer clusters. The inherent long lasting, repeated, and random file access pattern can easily saturate the metadata and data service and negatively impact other users. In this paper, we present FanStore, a transient runtime file system that optimizes DL I/O on existing hardware/software stacks. FanStore distributes datasets to the local storage of compute nodes, and maintains a global namespace. With the techniques of system call interception, distributed metadata management, and generic data compression, FanStore provides a POSIX-compliant interface with native hardware throughput in an efficient and scalable manner. Users do not have to make intrusive code changes to use FanStore and take advantage of the optimized I/O. Our experiments with benchmarks and real applications show that FanStore can scale DL training to 512 compute nodes with over 90\% scaling efficiency. …