BEGIN:VCALENDAR
VERSION:2.0
PRODID:IEEE vTools.Events//EN
CALSCALE:GREGORIAN
BEGIN:VTIMEZONE
TZID:America/Chicago
BEGIN:DAYLIGHT
DTSTART:20260308T030000
TZOFFSETFROM:-0600
TZOFFSETTO:-0500
RRULE:FREQ=YEARLY;BYDAY=2SU;BYMONTH=3
TZNAME:CDT
END:DAYLIGHT
BEGIN:STANDARD
DTSTART:20251102T010000
TZOFFSETFROM:-0500
TZOFFSETTO:-0600
RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=11
TZNAME:CST
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260210T220010Z
UID:044F7FF9-7F23-4766-8097-CB8C43B56196
DTSTART;TZID=America/Chicago:20260112T180000
DTEND;TZID=America/Chicago:20260112T200000
DESCRIPTION:The talk will address three-pillars to sustainable AI:\n\nEnerg
 y &amp; Infrastructure: Leveraging cloud computing&#39;s superior energy efficienc
 y compared to on-premises solutions and selecting regions with renewable e
 nergy sources.\n\nModel Optimization: Choosing appropriately-sized models 
 for specific tasks and implementing techniques like model distillation\, q
 uantization\, pruning\, and prompt caching to reduce resource consumption 
 while maintaining performance.\n\nFrugality: Questioning whether AI is nec
 essary for specific problems and considering less resource-intensive alter
 natives.\n\nThe presentation emphasizes incorporating sustainability as a 
 core requirement alongside cost and security considerations in AI developm
 ent.\n\nSpeaker(s): Sowjanya Pandruju\, \n\nAgenda: \n\n\nRoom: 302\, Bldg
 : Madison Central Library\, 201 West Mifflin Street\, Madison\, Wisconsin\
 , United States\, 53703
LOCATION:Room: 302\, Bldg: Madison Central Library\, 201 West Mifflin Stree
 t\, Madison\, Wisconsin\, United States\, 53703
ORGANIZER:harrison@glsan.com
SEQUENCE:8
SUMMARY:Optimizing AI for Sustainability
URL;VALUE=URI:https://events.vtools.ieee.org/m/531614
X-ALT-DESC:Description: &lt;br /&gt;&lt;p&gt;The talk will address three-pillars to sus
 tainable AI:&lt;br&gt;&lt;br&gt;Energy &amp;amp\; Infrastructure: Leveraging cloud computi
 ng&#39;s superior energy efficiency compared to on-premises solutions and sele
 cting regions with renewable energy sources.&lt;br&gt;&lt;br&gt;Model Optimization: Ch
 oosing appropriately-sized models for specific tasks and implementing tech
 niques like model distillation\, quantization\, pruning\, and prompt cachi
 ng to reduce resource consumption while maintaining performance.&lt;br&gt;&lt;br&gt;Fr
 ugality: Questioning whether AI is necessary for specific problems and con
 sidering less resource-intensive alternatives.&lt;br&gt;&lt;br&gt;The presentation emp
 hasizes incorporating sustainability as a core requirement alongside cost 
 and security considerations in AI development.&lt;/p&gt;\n&lt;div id=&quot;fortidlp-inpu
 t-element-parent&quot;&gt;&lt;/div&gt;&lt;br /&gt;&lt;br /&gt;Agenda: &lt;br /&gt;&lt;p&gt;&amp;nbsp\;&lt;/p&gt;\n&lt;div id=
 &quot;fortidlp-input-element-parent&quot;&gt;&lt;/div&gt;
END:VEVENT
END:VCALENDAR