Takashi Chikayama

A Portable and Efficient Implementation of KL1

A portable implementation scheme of a concurrent logic programming language KL1 by compiling it into C language code is investigated. A feature called generic objects was introduced to the scheme, which allows adding new data types and their manipulation without even slightly changing the core implementation. Parallel implementations can also be built upon it.

A scalable and efficient self-organizing failure detector for grid applications

Failure detection and group membership management are basic building blocks for self-repairing systems in distributed environments, which need to be scalable, reliable, and efficient in practice. As available resources become larger in size and more widely distributed, it is more essential that they can be easily used with a small amount of manual configuration in grid environments, where connectivities between different networks may be limited by firewalls and NATs. In this paper, we present a scalable failure detection protocol that self-organizes in grid environments. Our failure detectors autonomously create dispersed monitoring relationships among participating processes with almost no manual configuration so that each process will be monitored by a small number of other processes, and quickly disseminate notifications along the monitoring relationships when failures are detected. With simulations and real experiments, we showed that our failure detector has a practical scalability, a high reliability, and a good efficiency. The overhead with 313 processes was at most 2-percent even when the heartbeat interval was set to 0.1 second, and accordingly smaller when it was longer.

A'UM—a stream-based concurrent object-oriented language

This paper presents a computation model and its programming language,A’UM,* as a result of our pursuit of high parallelism and high expressivity for the development of a large scale software. By basing it on streams and integrating it with objects and relations,A’UM realizes an elegant model, natural representation and efficient execution, all at once, that have never been done by any other approaches.

ESP — Extended Self-contained PROLOG — as a preliminary Kernel Language of fifth generation computers

In the first three-year development stage of the fifth generation computer systems project, a series of high-performance personal computers called sequential inference machines are being developed at ICOT Research Center. The machines have a high-level machine language called KL0, which is a PROLOG-based logic language with various extensions. In the software development of the sequential inference machines. ESP, a software-supported yet higher level language compiled into KL0, is used instead of directly using KL0. This paper describes the design of the language system in sequential inference machines. This description will center on ESP with an overview of KL0.

A load balancing mechanism for large scale multiprocessor systems and its implementation

In large scale multiprocessor systems, the distance between processors should be taken into account by software to reduce the network traffic and the communication overhead. A load balancing method based on P3 (Processing Power Plane) model is proposed to enable programmers to specify distributing computational load, keeping the locality of the computation. In this method, a process is allocated to a rectangle on a hypothetical processing power plane. The size of the rectangle represents the processing power given to the process, and the distance between rectangles represents the communication cost between them. This plane is divided to processors, and the region of the processor may be dynamically reshaped to alleviate imbalance on P3. Mechanism for realization of the method has been implemented on the Multi-PSI/version, 2, which is a parallel processing system with 64 processing elements connected to form a 2-dimensional mesh network. A packet transmission mechanism of the Multi-PSI/version 2 is described, which realizes the process distribution along with the balancing method.

Distributed memory implementation of KLIC

This paper describes external reference management and distributed unification in a distributed implementation of a concurrent logic programming language KL1. This implementation is based on the KLIC system. KLIC has a feature calledgeneric objects that enable easy modification and extension of the system without changes in the core implementation. This distributed implementation is built upon the same core and external references are represented using generic objects. Unification operations are defined as methods of generic objects. Since creation of interprocessor reference loops cannot be avoided, we studied a new unification scheme that can cope with interprocessor reference loops. We built several experimental distributed systems that all demonstrate reasonable efficiency.

A Fast and Accurate Method for Detecting English-Japanese Parallel Texts

Parallel corpus is a valuable resource used in various fields of multilingual natural language processing. One of the most significant problems in using parallel corpora is the lack of their availability. Researchers have investigated approaches to collecting parallel texts from the Web. A basic component of these approaches is an algorithm that judges whether a pair of texts is parallel or not. In this paper, we propose an algorithm that accelerates this task without losing accuracy by preprocessing a bilingual dictionary as well as the collection of texts. This method achieved 250,000 pairs/sec throughput on a single CPU, with the best F1 score of 0.960 for the task of detecting 200 Japanese-English translation pairs out of 40,000. The method is applicable to texts of any format, and not specific to HTML documents labeled with URLs. We report details of these preprocessing methods and the fast comparison algorithm. To the best of our knowledge, this is the first reported experiment of extracting Japanese-English parallel texts from a large corpora based solely on linguistic content.

Collective operations for wide-area message passing systems using adaptive spanning trees

We propose a method for wide-area message passing systems to perform collective operations using dynamically created spanning trees. In our proposal, broadcasts and reductions are performed efficiently using topology-aware spanning trees constructed at run-time; processors autonomously measure latency and bandwidth to create latency-aware trees for short messages and bandwidth-aware trees for long messages. Our spanning trees adapt to topology changes due to the joining or leaving of processors; when processors join or leave a computation, processors repair the spanning trees so that effective execution of collective operations can continue. With 128 to 201 processors distributed over 3 to 4 clusters, the latency of our broadcast was within a factor of 2 of a static topology-aware implementation, and our broadcast achieved 82 percent of the bandwidth of a static topology-aware implementation. Moreover, when some processors joined or left a computation, our broadcast temporarily performed poorly for about 8 seconds while the spanning trees adapted to the new topology, but completed successfully even during this time.

Wide-coverage relation extraction from MEDLINE using deep syntax.

BackgroundRelation extraction is a fundamental technology in biomedical text mining. Most of the previous studies on relation extraction from biomedical literature have focused on specific or predefined types of relations, which inherently limits the types of the extracted relations. With the aim of fully leveraging the knowledge described in the literature, we address much broader types of semantic relations using a single extraction framework.ResultsOur system, which we name PASMED, extracts diverse types of binary relations from biomedical literature using deep syntactic patterns. Our experimental results demonstrate that it achieves a level of recall considerably higher than the state of the art, while maintaining reasonable precision. We have then applied PASMED to the whole MEDLINE corpus and extracted more than 137 million semantic relations. The extracted relations provide a quantitative understanding of what kinds of semantic relations are actually described in MEDLINE and can be ultimately extracted by (possibly type-specific) relation extraction systems.ConclusionPASMED extracts a large number of relations that have previously been missed by existing text mining systems. The entire collection of the relations extracted from MEDLINE is publicly available in machine-readable form, so that it can serve as a potential knowledge base for high-level text-mining applications.

Comparison Training of Shogi Evaluation Functions with Self-Generated Training Positions and Moves

Automated tuning of parameters in computer game playing is an important technique for building strong computer programs. Comparison training is a supervised learning method for tuning the parameters of an evaluation function. It has proven to be effective in the game of Chess and Shogi. The training method requires a large number of training positions and moves extracted from game records of human experts; however, the number of such game records is limited. In this paper, we propose a practical approach to create additional training data for comparison training by using the program itself. We investigate three methods for generating additional positions and moves. Then we evaluate them using a Shogi program. Experimental results show that the self-generated training data can improve the playing strength of the program.