- Chair: Yoshiaki Kakuda (Hiroshima City University)
- Speaker: Hidenori Nakazato (Waseda University)
- Title: Federating Autonomous IoT Silos: Fed4IoT Approach
- Many IoT systems are being installed to make human life easier. Each IoT system is often to be a silo which is isolated from other systems and is required to be deployed ground up. It causes high initial cost to develop IoT services. Fed4IoT, an EU-JP joint project, is initiated to amend this problem. The goal of the project is to federate isolated IoT systems, to make IoT devices sharable among many IoT services over many IoT silos, and to promote IoT service development. This talk introduces the key concept in the project, the architecture of the system, and the current status of the project.
- Chairs: Kentaro Sano (RIKEN), Michihiro Koibuchi (National Institute of Informatics)
- Speaker: Masato Motomura (Tokyo Institute of Technology)
- Title: Domain-Specific Architectures for Boosting “Compute for Intelligence”
- Abstract: Hardware-software joint research based on domain-knowledge is getting more crucial as “AI applications” spread worldwide as primary workloads to compute. The talk will cover such examples in deep neural networks, ensemble learning, and discrete optimizations.
- Chair: Takeshi Nanri(Kyushu University)
- Speaker Keiichiro Fukazawa (Academic Center for Computing and Media Studies, Kyoto University)
- Title: Research of applying the machine learning to space plasma physics with observation and numerical simulation data
- Abstract: The machine learning has become a powerful tool to find the relation between variables thanks to the deep learning technique. This performs greatly in the classification, regression and recently generation in the engineering and commercial areas. However, due to the satisfaction of physical laws in the scientific research area, the application of machine learning has some difficulties. In particular, the generation is very sensitive to scientific data since the generated data is not guaranteed by the physical laws. To overcome these problems, we have started the research how to apply machine learning to space plasma physics. In this talk, we show difficulties and efforts of applying the machine learning to the physical science.
- Chair: Nobuhiko Nakano (Keio University)
- Speaker: Anh Vu Doan (Technical University of Munich)
- Title: Statistical and Stochastic Methodologies for Fine-Grained Power Modeling of Multicore Processors
- Abstract: To take effective management decisions, both power and thermal management of multicores depend on accurate run-time dynamic power consumption information at core-level. Due to the cost-prohibitive nature of actually measuring core power, such run-time power information is usually derived from predetermined power models which use observable performance counters, operating frequency and voltage as inputs. The performance counters are necessary to model the activity and thus indirectly the power consumption of each core. Deriving such power information is mainly based on predetermined power models which use linear modeling techniques to determine the core-performance/core-power relationship. However, with multicore processors becoming ever more complex, linear modeling techniques cannot capture all possible core-performance related power states anymore. In this talk, we present different statistical and stochastic (including neural networks) methodologies that can be used to better model power consumption at core-level, hence allowing to improve power and thermal management for more sustainable embedded systems.