Efficient and reliable methods for training of object detectors are in higher demand than ever, and more and more data relevant to the field is becoming available. However, large datasets like Open Images Dataset v4 (OID) are sparsely annotated, and some measure must be taken in order to ensure the training of a reliable detector. In order to take the incompleteness of these datasets into account, one possibility is to use pretrained models to detect the presence of the unverified objects. However, the performance of such a strategy depends largely on the power of the pretrained model. In this study, we propose part-aware sampling, a method that uses human intuition for the hierarchical relation between objects. In terse terms, our method works by making assumptions like “a bounding box for a car should contain a bounding box for a tire”. We demonstrate the power of our method on OID and compare the performance against a method based on a pretrained model. Our method also won the first and second place on the public and private test sets of the Google AI Open Images Competition 2018.

Software frameworks for neural networks play a key role in the development and application of deep learning methods. In this paper, we introduce the Chainer framework, which intends to provide a flexible, intuitive, and high performance means of implementing the full range of deep learning models needed by researchers and practitioners. Chainer provides acceleration using Graphics Processing Units with a familiar NumPy-like API through CuPy, supports general and dynamic models in Python through Define-by-Run, and also provides add-on packages for state-of-the-art computer vision models as well as distributed training.

An important concern for current IaaS cloud providers is to be falsely accused by cloud users for the response delays of their applications running on the cloud. Since cloud computing brings an additional virtualization layer between guest-OS and hardware resources, the relationship between the performance of an application and its resource consumption becomes obscure. Only monitoring resource consumption (e.g., CPU, memory, disk I/O and so on) is often inadequate to detect the QoS degradation of a deployed application, not to mention resolving cloud users' QoS complaints. In this paper, we introduce a novel single-view performance visualization technique for a cloud platform to monitor the response delays of all the on-line applications running on it. Our technique sniffs network packets from a network switch in the platform and analyzes their flow to estimate the response time of each application. An important merit of this approach is the easy installation into a cloud platform. Concretely, our approach only requires changing the configuration of a network switch to enable port mirroring. To detect and visualize the response delay of each application in a single view, we convert the absolute response time of each application to a comparable value of delay degree ("delay ratio") which shows how serious the response delays compared to its baseline. The baseline is the service time of each application, which we estimate from the distribution of observed response times (service time + waiting time) by adopting an exponentially modified Gaussian (EMG) distribution fitting with an outlier elimination. We validate our approach through extensive experiments of a representative on-line application benchmark (RUBBoS). The results show high repeatability of the service time estimation (the standard deviation of 40 trials is 0.11). The results also show that our performance monitoring technique is able to accurately visualize response delays even when a training set of response times has a heavy-tailed distribution.

MOTIVATION: Sequence homology searches are used in various fields. New sequencing technologies produce huge amounts of sequence data, which continuously increase the size of sequence databases. As a result, homology searches require large amounts of computational time, especially for metagenomic analysis. RESULTS: We developed a fast homology search method based on database subsequence clustering, and implemented it as GHOSTZ. This method clusters similar subsequences from a database to perform an efficient seed search and ungapped extension by reducing alignment candidates based on triangle inequality. The database subsequence clustering technique achieved an ∼2-fold increase in speed without a large decrease in search sensitivity. When we measured with metagenomic data, GHOSTZ is ∼2.2-2.8 times faster than RAPSearch and is ∼185-261 times faster than BLASTX. AVAILABILITY AND IMPLEMENTATION: The source code is freely available for download at http://www.bi.cs.titech.ac.jp/ghostz/ CONTACT: [email protected] SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

SUMMARY: The application of protein-protein docking in large-scale interactome analysis is a major challenge in structural bioinformatics and requires huge computing resources. In this work, we present MEGADOCK 4.0, an FFT-based docking software that makes extensive use of recent heterogeneous supercomputers and shows powerful, scalable performance of >97% strong scaling. AVAILABILITY AND IMPLEMENTATION: MEGADOCK 4.0 is written in C++ with OpenMPI and NVIDIA CUDA 5.0 (or later) and is freely available to all academic and non-profit users at: http://www.bi.cs.titech.ac.jp/megadock. CONTACT: [email protected] SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Many structured overlays forward messages based not only on node identifiers but also on the other factors such as network proximity. It is an essential requirement for a visualization system to exhibit how the factors are considered in a overlay. Such a visualization system allows us to grasp intuitively characteristics of an overlay algorithm. If this is achieved, differences between algorithms become more distinguishable, and those who are unfamiliar with an algorithm are also able to know improvement between algorithms. Thus, such a visualization increases their interests in research activity in peer-to-peer computing. In this demonstration we propose a novel 3D visualization system for structured overlays. The system is practically able to visualize how routing algorithms consider node groups in an effective manner. We demonstrate the system with Chord and GFRT-Chord algorithms, but the system is able to visualize algorithms not limited to them.

The optical cross-connect (OXC) system described in this paper increases the operation flexibility and reliability of the trunk-line optical networks used for data communication. It features an OXC node comprised of a photonic switching network and a conventional electric switching network that are connected hierarchically. The operation and management scheme proposed for this OXC system uses the concept of an optical path and an optical section. The OXC system allows hitless network reconfiguration by switching the photonic switches gradually and without interference effects. An experimental OXC network showed that a broken optical path is restored, by rerouting, within 50 ms. Experiments using a LiNbO/sub 3/ 8/spl times/8 photonic switch matrix also showed that the OXC system provides photonic hitless switching. These results confirm the feasibility of flexibly reconfigurable and fast-restorable OXC systems.