In this paper, we propose a novel edge-labeling graph neural network (EGNN), which adapts a deep neural network on the edge-labeling graph, for few-shot learning. The previous graph neural network (GNN) approaches in few-shot learning have been based on the node-labeling framework, which implicitly models the intra-cluster similarity and the inter-cluster dissimilarity. In contrast, the proposed EGNN learns to predict the edge-labels rather than the node-labels on the graph that enables the evolution of an explicit clustering by iteratively updating the edgelabels with direct exploitation of both intra-cluster similarity and the inter-cluster dissimilarity. It is also well suited for performing on various numbers of classes without retraining, and can be easily extended to perform a transductive inference. The parameters of the EGNN are learned by episodic training with an edge-labeling loss to obtain a well-generalizable model for unseen low-data problem. On both of the supervised and semi-supervised few-shot image classification tasks with two benchmark datasets, the proposed EGNN significantly improves the performances over the existing GNNs.

In this paper, a neural architecture search (NAS) framework is proposed for 3D medical image segmentation, to automatically optimize a neural architecture from a large design space. Our NAS framework searches the structure of each layer including neural connectivities and operation types in both of the encoder and decoder. Since optimizing over a large discrete architecture space is difficult due to high-resolution 3D medical images, a novel stochastic sampling algorithm based on a continuous relaxation is also proposed for scalable gradient based optimization. On the 3D medical image segmentation tasks with a benchmark dataset, an automatically designed architecture by the proposed NAS framework outperforms the human-designed 3D U-Net, and moreover this optimized architecture is well suited to be transferred for different tasks.

Layer normalization is a recently introduced technique for normalizing the activities of neurons in deep neural networks to improve the training speed and stability. In this paper, we introduce a new layer normalization technique called Dynamic Layer Normalization (DLN) for adaptive neural acoustic modeling in speech recognition. By dynamically generating the scaling and shifting parameters in layer normalization, DLN adapts neural acoustic models to the acoustic variability arising from various factors such as speakers, channel noises, and environments. Unlike other adaptive acoustic models, our proposed approach does not require additional adaptation data or speaker information such as i-vectors. Moreover, the model size is fixed as it dynamically generates adaptation parameters. We apply our proposed DLN to deep bidirectional LSTM acoustic models and evaluate them on two benchmark datasets for large vocabulary ASR experiments: WSJ and TED-LIUM release 2. The experimental results show that our DLN improves neural acoustic models in terms of transcription accuracy by dynamically adapting to various speakers and environments.

No abstract

Due to the inherent model uncertainty, learning to infer Bayesian posterior from a few-shot dataset is an important step towards robust meta-learning. In this paper, we propose a novel Bayesian model-agnostic meta-learning method. The proposed method combines efficient gradient-based meta-learning with nonparametric variational inference in a principled probabilistic framework. Unlike previous methods, during fast adaptation, the method is capable of learning complex uncertainty structure beyond a simple Gaussian approximation, and during meta-update, a novel Bayesian mechanism prevents meta-level overfitting. Remaining a gradientbased method, it is also the first Bayesian model-agnostic meta-learning method applicable to various tasks including reinforcement learning. Experiment results show the accuracy and robustness of the proposed method in sinusoidal regression, image classification, active learning, and reinforcement learning.

This paper addresses the problem of efficiently solving large-scale energy minimization problems encountered in computer vision. We propose an energy-aware method for merging random variables to reduce the size of the energy to be minimized. The method examines the energy function to find groups of variables which are likely to take the same label in the minimum energy state and thus can be represented by a single random variable. We propose and evaluate a number of extremely efficient variable grouping strategies. Experimental results show that our methods result in a dramatic reduction in the computational cost and memory requirements (in some cases by a factor of one hundred) with almost no drop in the accuracy of the final result. Comparative evaluation with efficient super-pixel generation methods, which are commonly used in variable grouping, reveals that our methods are far superior both in terms of accuracy and running time.

PurposeThis study was performed to assess potential improvements in clinical outcomes when applying recent advanced hip arthroplasty surgical techniques and understand the potential relationship between bone mineral density (BMD) and surgical outcomes.Materials and MethodsAmong 37 cases of periprosthetic femoral fractures after hip arthroplasty treated between March 2014 and September 2016, all included a follow-up of at least one year and were included in this study. Outcomes were evaluated using the Beals and Tower's criteria. BMD was examined in 27 of 37 cases and the relationship between osteoporosis and treatment outcomes was analyzed. Advanced hip arthroplasty surgical approaches varied depending on the fracture type: i) open reduction with wiring for Vancouver A, ii) open reduction with double plate fixation for Vancouver B1, iii) revision THA with long stem for Vancouver B1-Nonunion, B2 and B3, and iv) open reduction with double plate fixation for Vancouver C.ResultsWhen assessed using the Beals and Tower's criteria, 33 out of 37 (89.2%) patients were excellent and 4 (10.8%) were poor. These outcomes were an improvement compared with series I (81.8%). When analyzed according to the Vancouver classification, patients with type A (n=8), type B1 (n=16), and type B2 (n=2) were all excellent, the patients with type B3 were excellent (n=1) and poor (n=1), and the patients with type C were excellent (n=6) and poor (n=3). The mean BMD was −2.6 (T-score) in 27 of 37 cases and −4.4 in 4 cases with poor prognosis. Osteoporosis was statistically correlated to those classified as poor by Beals and Tower.ConclusionThe results of the analysis suggest that applying new surgical hip arthroplasty treatment approaches leads to improved outcomes compared with the author's previous study. When treating periprosthetic femoral fractures following total hip arthroplasty, an appropriate internal fixation method should be selected, at least in part based on the Vancouver classification. In addition, osteoporosis may be a major prognostic factor for the outcomes of surgical treatment.

No abstract