Bright Spark Lectures


Bright Spark Lecture 3 Sparse Modeling in Geotechnical Engineering
Date / Time 13 December 2019, Friday / 16:30 - 18:00 hrs
Venue Room IB-101
Speaker Takayuki Shuku
Wuhan University, China

Biography

Takayuki Shuku, is an Associate Professor at Okayama University, Japan.He obtained his BE and MS from Shimane University and his PhD degree from Okayama University. His main research interests include inverse analysis and machine learning in geotechnical engineering. He received the best paper award and the best young researcher award by the Japanese Geotechnical Society in2013, the award for among the top five downloaded articles in Granular Matter during 2018.

Abstract

Estimation and prediction problems in geotechnical engineering belong to the class of inverse problems. Many different approaches to analyze these problems have been reported in the literature. In recent years, the use of machine learning has become increasingly common in many research fields because of the rapid increase of computational capacity and advances in algorithms. In particular, the methodology for solving inverse problems known as “sparse modeling” has been receiving considerable attention. Sparse modeling is a statistical method which exploits specific features/structures in data based on solution sparsity, and has a great potential for application to geotechnical problems. This paper demonstrates the potential of sparse modeling for solving geotechnical engineering problems by means of two practical examples: a crossborehole tomography, and a stratigraphic soil profiling based on cone penetration test.


Full Text (PDF)