Date: July 19, 2019
Source: State Key Laboratory of Acoustics, IACAS
Invited by the State Key Laboratory of Acoustics, Prof. Peter Gerstoft from Scripps Institution of Oceanography of University of California San Diego visited the Institute of Acoustics of Chinese Academy of Sciences (IACAS) on July 19, 2019. He gave an academic lecture entitled Machine learning and applications in ocean acoustics in IACAS. More than sixty staff and students attended this seminar.
Prof. Peter Gerstoft briefly introduced the data-driven machine learning methods in ocean acoustics and the difference from other conventional model-based approaches at first. Then he gave a detailed presentation on the source localization and Direction-Of-Arrival (DOA) estimation using supervised machine learning including neural networks and compressive sensing.
Prof. Peter Gerstoft showed several experimental results on source localization and DOA estimation to demonstrate the prospect of machine learning in ocean acoustics. He also described the applications of unsupervised machine learning methods, including graph signal processing for source localization and dictionary learning for travel time tomography. At the end of the lecture, he emphasized that there were lots of research opportunities for machine learning in ocean acoustics and geophysics.
Prof. Peter Gerstoft received his Ph.D. from the Technical University of Denmark in 1986. From 1987-1997, he worked at Massachusetts Institute of Technology, Woods Hole Oceanographic Institute and SACLANT Undersea Research Centre successively. Since 1997, he has been a professor in Scripps Institution of Oceanography, University of California San Diego. He has published over 200 papers with more than 8200 citations (Google Scholar). Prof. Peter Gerstoft is also a Fellow of the Acoustical Society of America (ASA) and the chair of the 178th ASA Meeting to be held on December 2-6, 2019.
Prof. Peter Gerstoft is presenting his report (Image by IACAS)