BEGIN:VCALENDAR
VERSION:2.0
PRODID:IEEE vTools.Events//EN
CALSCALE:GREGORIAN
BEGIN:VTIMEZONE
TZID:America/Los_Angeles
BEGIN:DAYLIGHT
DTSTART:20260308T030000
TZOFFSETFROM:-0800
TZOFFSETTO:-0700
RRULE:FREQ=YEARLY;BYDAY=2SU;BYMONTH=3
TZNAME:PDT
END:DAYLIGHT
BEGIN:STANDARD
DTSTART:20261101T010000
TZOFFSETFROM:-0700
TZOFFSETTO:-0800
RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=11
TZNAME:PST
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260811T042101Z
UID:1BED1507-ED28-45E4-A985-BD324962D244
DTSTART;TZID=America/Los_Angeles:20260730T120000
DTEND;TZID=America/Los_Angeles:20260730T130000
DESCRIPTION:Advanced synthesis methods have made it possible to create mill
 ions of nanoparticles on a single chip. These megalibraries provide a weal
 th of information\, but their formidable size renders manual characterizat
 ion infeasible. This talk will describe the development of novel artificia
 l intelligence techniques to automate nanoparticle characterization. First
 \, we show that neural architecture search can create machine learning mod
 els that generalize well to previously-unseen nanoparticle datasets. Next\
 , we demonstrate a rapid\, unsupervised segmentation technique for our dat
 a that is expected to perform at least 25 times faster\, on average\, than
  the benchmark method. We then describe how image processing can be applie
 d to nanoparticle images before machine learning model training such that 
 accuracy is improved by >10% relative to baseline. These artificial intell
 igence techniques are expected to significantly accelerate nanoparticle an
 alysis relative to manual processes and represent a step towards “self-d
 riving” laboratories.\n\nSpeaker(s): Prof. Alexandra Day\n\nBldg: 232\, 
 833 Dyer Road\, Monterey\, California\, United States\, 93940
LOCATION:Bldg: 232\, 833 Dyer Road\, Monterey\, California\, United States\
 , 93940
ORGANIZER:christopher.gaytan1@nps.edu
SEQUENCE:14
SUMMARY:Development of Artificial Intelligence Techniques for Automated Nan
 oparticle Characterization
URL;VALUE=URI:https://events.vtools.ieee.org/m/572121
X-ALT-DESC:Description: <br /><p><span style="color: rgb(0\, 0\, 0)\; font-
 family: Aptos\, Aptos_EmbeddedFont\, Aptos_MSFontService\, Calibri\, Helve
 tica\, sans-serif\; font-size: 16px\; font-style: normal\; font-variant-li
 gatures: normal\; font-variant-caps: normal\; font-weight: 400\; letter-sp
 acing: normal\; orphans: 2\; text-align: start\; text-indent: 0px\; text-t
 ransform: none\; widows: 2\; word-spacing: 0px\; -webkit-text-stroke-width
 : 0px\; white-space: normal\; background-color: rgb(255\, 255\, 255)\; tex
 t-decoration-thickness: initial\; text-decoration-style: initial\; text-de
 coration-color: initial\; display: inline !important\; float: none\;">Adva
 nced synthesis methods have made it possible to create millions of nanopar
 ticles on a single chip. &nbsp\;These megalibraries provide a wealth of in
 formation\, but their formidable size renders manual characterization infe
 asible. &nbsp\;This talk will describe the development of novel artificial
  intelligence techniques to automate nanoparticle characterization. &nbsp\
 ;First\, we show that neural architecture search can create machine learni
 ng models that generalize well to previously-unseen nanoparticle datasets.
  &nbsp\;Next\, we demonstrate a rapid\, unsupervised segmentation techniqu
 e for our data that is expected to perform at least 25 times faster\, on a
 verage\, than the benchmark method. &nbsp\;We then describe how image proc
 essing can be applied to nanoparticle images before machine learning model
  training such that accuracy is improved by &gt\;10% relative to baseline.
  &nbsp\;These artificial intelligence techniques are expected to significa
 ntly accelerate nanoparticle analysis relative to manual processes and rep
 resent a step towards &ldquo\;self-driving&rdquo\; laboratories.</span></p
 >
END:VEVENT
END:VCALENDAR