トピックスTopics

IPIN 2019で2本の論文がBest Paper Award、Best Student Paper Awardにノミネートされました
Two Papers are nominated as Best Paper Award and Best Student Paper Award at IPIN 2019

以下の2本の論文が、屋内位置情報の国際会議IPIN 2019 (Tenth International Conference on Indoor Positioning and Indoor Navigation)(外部サイト)のBest PaperおよびBest Student Paperにノミネートされました。採択された論文のうち、Top15%の論文がノミネートされます。

■Best Paper Candidate:
No-Sweat Detective: No Effort Anomaly Detection for Wi-Fi-Based Localization
Kota Tsubouchi, Kohei Yamamoto (Ritsumeikan University) and Nobuhiko Nishio (Ritsumeikan University)
ユーザーから得られるWi-Fiのfingerprint情報を取得し、屋内測位時のキャリブレーションの労力を最小限にすることを狙った技術です。

■Best Student Paper Candidate:
GroupWi-Lo: Maintaining Wi-Fi-based Indoor Localization Accurate via Group-wise Total Variation Regularization
Masato Sugasaki (Tokyo Institute of Technology), Kota Tsubouchi, Masamichi Shimosaka (Tokyo Institute of Technology) and Nobuhiko Nishio (Ritsumeikan University)
学習の制約化項を工夫することにより、Wi-Fiのfingerprintによるモデリングをよりノイズ耐性に強く、より省コスト化したものです。

The following papers are nominated as Best Paper Award and Best Student Paper Award at IPIN 2019 (Tenth International Conference on Indoor Positioning and Indoor Navigation)(external site).

■Best Paper Candidate:
No-Sweat Detective: No Effort Anomaly Detection for Wi-Fi-Based Localization
Kota Tsubouchi, Kohei Yamamoto (Ritsumeikan University) and Nobuhiko Nishio (Ritsumeikan University)

Abstract:
We propose a new approach that detects anomalous reference points to gain felicitous supervised datasets in order to prevent overfitting. Unsupervised datasets obtained from off-the-shelf mobile navigation applications, i.e., user logs uploaded from phones, are used. Our approach is implemented in a system we call “No-Sweat Detective”. The results of an experiment in a controlled environment demonstrate that NoSweat Detective can detect anomalies caused by environmental changes, and the results of a five-month experiment show that NoSweat Detective has redundancy against a complex open-space environment in the real world.


■Best Student Paper Candidate:
GroupWi-Lo: Maintaining Wi-Fi-based Indoor Localization Accurate via Group-wise Total Variation Regularization
Masato Sugasaki (Tokyo Institute of Technology), Kota Tsubouchi, Masamichi Shimosaka (Tokyo Institute of Technology) and Nobuhiko Nishio (Ritsumeikan University)

Abstract:
We propose a brand new retraining method, called GroupWi-Lo, that focuses on minimization of parameter variation with respect to the incremental surveys on fingerprint (i.e., calibration). Our method tries to keep the parameters of the previously trained model unchanged while minimizing the error on the dataset obtained in the last surveys. This formulation is helpful to keep robustness against overfitting from the limited size of the dataset per survey.