Online learning algorithms often have to operate in the presence of concept drift (i.e., the concepts to be learned can change with time). This paper presents a new categorization for concept drift, separating drifts according to different criteria into mutually exclusive and nonheterogeneous categories. Moreover, although ensembles of learning machines have been used to learn in the presence of concept drift, there has been no deep study of why they can be helpful for that and which of their features can contribute or not for that. As diversity is one of these features, we present a diversity analysis in the presence of different types of drifts. We show that, before the drift, ensembles with less diversity obtain lower test errors. On the other hand, it is a good strategy to maintain highly diverse ensembles to obtain lower test errors shortly after the drift independent on the type of drift, even though high diversity is more important for more severe drifts. Longer after the drift, high diversity becomes less important. Diversity by itself can help to reduce the initial increase in error caused by a drift, but does not provide the faster recovery from drifts in long-term.

/pdf/the-impact-of-diversity-on-online-ensemble-learning-in-the-5apps5tlop.pdf

The Impact of Diversity on Online Ensemble Learning in the Presence of Concept Drift

Motivation: The Basic Local Alignment Search Tool (BLAST) is one of the most widely used bioinformatics tools. The widespread impact of BLAST is reflected in over 53 000 citations that this software has received in the past two decades, and the use of the word ‘blast’ as a verb referring to biological sequence comparison. Any improvement in the execution speed of BLAST would be of great importance in the practice of bioinformatics, and facilitate coping with ever increasing sizes of biomolecular databases.

Results: Using a general-purpose graphics processing unit (GPU), we have developed GPU-BLAST, an accelerated version of the popular NCBI-BLAST. The implementation is based on the source code of NCBI-BLAST, thus maintaining the same input and output interface while producing identical results. In comparison to the sequential NCBI-BLAST, the speedups achieved by GPU-BLAST range mostly between 3 and 4.

Availability: The source code of GPU-BLAST is freely available at http://archimedes.cheme.cmu.edu/biosoftware.html.

Contact: sahinidis@cmu.edu

Supplementary information:Supplementary data are available at Bioinformatics online.

Gpu-blast

DDoS attacks in cloud computing

URL: A unified reinforcement learning approach for autonomic cloud management

Data-intensive file systems, developed for Internet services and popular in cloud computing, provide high reliability and availability by replicating data, typically three copies of everything. Alternatively high performance computing, which has comparable scale, and smaller scale enterprise storage systems get similar tolerance for multiple failures from lower overhead erasure encoding, or RAID, organizations. DiskReduce is a modification of the Hadoop distributed file system (HDFS) enabling asynchronous compression of initially triplicated data down to RAID-class redundancy overheads. In addition to increasing a cluster's storage capacity as seen by its users by up to a factor of three, DiskReduce can delay encoding long enough to deliver the performance benefits of multiple data copies.

/pdf/diskreduce-raid-for-data-intensive-scalable-computing-27m36uidg8.pdf

DiskReduce: RAID for data-intensive scalable computing

Parallel and distributed systems

Strategies for implementing distance learning coursework have evolved and expanded with the growth and maturation of the World Wide Web. The requirements, advantages, and disadvantages of the most common strategies are compared and contrasted. Initially limited to individual efforts, software development has eased the burden of individual faculty and has opened up strategies for greater participation. Institutions attempting large‐scale implementations, however, may find infrastructure requirements overwhelming. An emerging large‐scale strategy involves an institutional partnership with a for‐profit application service provider (ASP). The ASP specializes in total systems solutions for developing and delivering Web‐based distance learning programs. Recent experiences at California State University, Fullerton, with an ASP are discussed.

Evolution of Web‐based distance learning strategies

Multimedia-Specific Technologies (DVI, CD-I, JPEG, MPEG) Multimedia Platforms and Peripherals Multimedia and Communicationns (WAN, LAN, ISDN, Fibre Optics, Satellites) Graphics, Audio and Video Hardware and Software Technologies Operating Systems (Windows, OS/2, UNIX) Authoring Tools and Languages Design and Development of MMC Media Preparation and Integration Interface Design End User Performance MMC and Information Systems MMC Training and Education Issues Commercial, Home, Entertainment Applications of MMC Issues of MMC Concerned with Legal, Copyright and Standards Matters Related Emerging Technologies (HDTV, Parallel Processing, Virtual Systems, Hypermedia) Global Issues of Development, Distribution and Manufacturing of MMC.

Multimedia Computing Preparing for the 21st Century

In recent years, there has been considerable interest in setting up technology-focused communities of practice for teaching and learning purposes. Instructors and students can play a key role in these communities of learning by creating and/or using Open Educational Resources (OERs). In 2016 two new discipline-specific communities were created within the MERLOT OER user community: Information Systems and Information Technology (IT/IS), and Computer Science (CS). The OER taxonomies for these, predicated on the CSAB/ABET curricula were a key factor in the endorsement of these communities and their portals, by the IEEE Computer Society and the IEEE Education Society. This paper describes the process of creation of both communities with the aim of establishing Editorial Boards to peer review OERs in these disciplines. Moreover, some analytics of the recently-created IT Editorial Board in MERLOT are presented as well.

Promoting MERLOT Communities Based on OERs in Computer Science and Information Systems

This study compares the effectiveness of classroom instruction and multimedia personal computing MPC instruction to teach undergraduate business school majors the principles of Lotus 1-2-3, WordPerfect 5.0, and dBase III+. The effectiveness of three interactive videodisc courses was compared to the effectiveness of regular, traditional, lecture classes TI. For this study, two groups of approximately 55 students each, were randomly selected from 15 sections of an introductory, undergraduate, business school program. For all three courses the posttest scores of the MPC students were higher than those of the TI students; in two of those cases, the differences were statistically significantly higher. In addition, there was no difference in the total average learning time for both groups. Students in the MPC group, surveyed about their learning experience, indicated exceptionally positive attitudes towards the courses they had used and to MPC in general.

Sorel Reisman

Papers

Parallel and distributed systems

Evolution of Web‐based distance learning strategies

Multimedia Computing Preparing for the 21st Century

Promoting MERLOT Communities Based on OERs in Computer Science and Information Systems

A Comparative Study of Multimedia Personal Computing and Traditional Instruction in a Business School Curriculum