BEGIN:VCALENDAR
VERSION:2.0
PRODID:IEEE vTools.Events//EN
CALSCALE:GREGORIAN
BEGIN:VTIMEZONE
TZID:Australia/Brisbane
BEGIN:STANDARD
DTSTART:19920301T020000
TZOFFSETFROM:+1100
TZOFFSETTO:+1000
TZNAME:AEST
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20250926T221705Z
UID:A9C45661-C716-4C4C-A3D3-75B48CB58670
DTSTART;TZID=Australia/Brisbane:20250806T150000
DTEND;TZID=Australia/Brisbane:20250806T160000
DESCRIPTION:Abstract: In recent years\, machine learning (ML) has garnered 
 significant attention for tackling challenges in wave imaging and sensing.
  However\, many approaches treat ML as a black box\, overlooking decades o
 f insight rooted in wave physics and rigorous mathematical analysis. This 
 talk highlights the importance of thoroughly understanding the forward pro
 blem that underpins these inherently inverse tasks\, and demonstrates how 
 integrating mathematical\, physical\, and engineering intuition can lead t
 o more efficient and elegant solutions.\n\nWe will explore millimeter-wave
  multiple-input multiple-output (MIMO) imaging\, showcasing how physics-in
 formed ML enhances reconstruction quality. Next\, we present a high-accura
 cy\, efficient ML classifier for 77 GHz FMCW radar. Rather than using raw 
 data\, it operates on a low-dimensional\, physics-derived input to classif
 y road targets into five categories\, achieving performance competitive wi
 th state-of-the-art methods in automotive radar applications.\n\nFinally\,
  we reflect on key lessons learned across these wave sensing and imaging c
 hallenges.\n\nSpeaker(s): Prof. Xudong Chen\, \n\nRoom: 914\, Bldg: 46 (An
 drew N. Liveris Building)\, The University of Queensland\, St Lucia\, Bris
 bane\, Queensland\, Australia\, 4072
LOCATION:Room: 914\, Bldg: 46 (Andrew N. Liveris Building)\, The University
  of Queensland\, St Lucia\, Brisbane\, Queensland\, Australia\, 4072
ORGANIZER:h.espinosa@griffith.edu.au
SEQUENCE:21
SUMMARY:Physics-Assisted Machine Learning for Wave Sensing and Imaging
URL;VALUE=URI:https://events.vtools.ieee.org/m/495464
X-ALT-DESC:Description: &lt;br /&gt;&lt;p&gt;&lt;img style=&quot;float: right\;&quot; src=&quot;https://e
 vents.vtools.ieee.org/vtools_ui/media/display/46a25ac8-d5a0-44c1-b03a-028c
 6eed8697&quot; width=&quot;210&quot; height=&quot;286&quot;&gt;&lt;/p&gt;\n&lt;p&gt;&lt;span style=&quot;font-size: 12pt\;
 &quot;&gt;&lt;strong&gt;Abstract&lt;/strong&gt;: In recent years\, machine learning (ML) has g
 arnered significant attention for tackling challenges in wave imaging and 
 sensing. However\, many approaches treat ML as a black box\, overlooking d
 ecades of insight rooted in wave physics and rigorous mathematical analysi
 s. This talk highlights the importance of thoroughly understanding the for
 ward problem that underpins these inherently inverse tasks\, and demonstra
 tes how integrating mathematical\, physical\, and engineering intuition ca
 n lead to more efficient and elegant solutions.&lt;/span&gt;&lt;/p&gt;\n&lt;p&gt;&lt;span style
 =&quot;font-size: 12pt\;&quot;&gt;We will explore millimeter-wave multiple-input multip
 le-output (MIMO) imaging\, showcasing how physics-informed ML enhances rec
 onstruction quality. Next\, we present a high-accuracy\, efficient ML clas
 sifier for 77 GHz FMCW radar. Rather than using raw data\, it operates on 
 a low-dimensional\, physics-derived input to classify road targets into fi
 ve categories\, achieving performance competitive with state-of-the-art me
 thods in automotive radar applications.&lt;/span&gt;&lt;/p&gt;\n&lt;p&gt;&lt;span style=&quot;font-s
 ize: 12pt\;&quot;&gt;Finally\, we reflect on key lessons learned across these wave
  sensing and imaging challenges.&lt;/span&gt;&lt;/p&gt;
END:VEVENT
END:VCALENDAR