Source-Guided Discrepancy (S-disc) google
Unsupervised domain adaptation is the problem setting where data generating distributions in the source and target domains are different, and labels in the target domain are unavailable. One important question in unsupervised domain adaptation is how to measure the difference between the source and target domains. A previously proposed discrepancy that does not use the source domain labels requires high computational cost to estimate and may lead to a loose generalization error bound in the target domain.To mitigate these problems, we propose a novel discrepancy called source-guided discrepancy ($S$-disc), which exploits labels in the source domain. As a consequence, $S$-disc can be computed efficiently with a finite sample convergence guarantee. In addition, we show that $S$-disc can provide a tighter generalization error bound than the one based on an existing discrepancy. Finally, we report experimental results that demonstrate the advantages of $S$-disc over the existing discrepancies. …

Spontaneous Clustering google
We propose a new method for clustering based on the local minimization of the \gamma-divergence, which we call the spontaneous clustering. The greatest advantage of the proposed method is that it automatically detects the number of clusters that adequately reflect the data structure. In contrast, exiting methods such as K-means, fuzzy c-means, and model based clustering need to prescribe the number of clusters. We detect all the local minimum points of the \gamma-divergence, which are defined as the centers of clusters. A necessary and sufficient condition for the \gamma-divergence to have the local minimum points is also derived in a simple setting. A simulation study and a real data analysis are performed to compare our proposal with existing methods. …

MetaPruning google
In this paper, we propose a novel meta learning approach for automatic channel pruning of very deep neural networks. We first train a PruningNet, a kind of meta network, which is able to generate weight parameters for any pruned structure given the target network. We use a simple stochastic structure sampling method for training the PruningNet. Then, we apply an evolutionary procedure to search for good-performing pruned networks. The search is highly efficient because the weights are directly generated by the trained PruningNet and we do not need any finetuning. With a single PruningNet trained for the target network, we can search for various Pruned Networks under different constraints with little human participation. We have demonstrated competitive performances on MobileNet V1/V2 networks, up to 9.0/9.9 higher ImageNet accuracy than V1/V2. Compared to the previous state-of-the-art AutoML-based pruning methods, like AMC and NetAdapt, we achieve higher or comparable accuracy under various conditions. …

MatRox google
We present MatRox, a novel model-based algorithm and implementation of Hierarchically Semi-Separable (HSS) matrix computations on parallel architectures. MatRox uses a novel storage format to improve data locality and scalability of HSS matrix-matrix multiplications on shared memory multicore processors. We build a performance model for HSS matrix-matrix multiplications. Based on the performance model, a mixed-rank heuristic is introduced to find an optimal HSS-tree depth for a faster HSS matrix evaluation. Uniform sampling is used to improve the performance of HSS compression. MatRox outperforms state-of-the-art HSS matrix multiplication codes, GOFMM and STRUMPACK, with average speedups of 2.8x and 6.1x respectively on target multicore processors. …