BEGIN:VCALENDAR VERSION:2.0 PRODID:IEEE vTools.Events//EN CALSCALE:GREGORIAN BEGIN:VTIMEZONE TZID:America/Los_Angeles BEGIN:DAYLIGHT DTSTART:20240310T030000 TZOFFSETFROM:-0800 TZOFFSETTO:-0700 RRULE:FREQ=YEARLY;BYDAY=2SU;BYMONTH=3 TZNAME:PDT END:DAYLIGHT BEGIN:STANDARD DTSTART:20241103T010000 TZOFFSETFROM:-0700 TZOFFSETTO:-0800 RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=11 TZNAME:PST END:STANDARD END:VTIMEZONE BEGIN:VEVENT DTSTAMP:20240424T044232Z UID:3ABADD18-D3B1-4A3F-BF1D-1C8D4A4871EA DTSTART;TZID=America/Los_Angeles:20240423T173000 DTEND;TZID=America/Los_Angeles:20240423T183000 DESCRIPTION:Despite the widespread proliferation of neural networks\, the m echanisms through which they operate so successfully are not well understo od. In this talk\, we will first explore empirical and theoretical investi gations into neural network training and generalization and what they can tell us about why deep learning works. Then\, we will examine a recent lin e of work on algorithm learning. While neural networks typically excel at pattern matching tasks\, we consider whether neural networks can learn alg orithms that scale to problem instances orders of magnitude larger than th ose seen during training.\n\nSpeaker(s): Micah Goldblum\n\nAgenda: \n- Inv ited talk from Micah Goldblum\, postdoctoral research fellow at New York U niversity working with [Yann LeCun](https://research.facebook.com/people/l ecun-yann/) and [Andrew Gordon Wilson](https://cims.nyu.edu/~andrewgw/).\n \n- Q/A Session\n\nVirtual: https://events.vtools.ieee.org/m/410297 LOCATION:Virtual: https://events.vtools.ieee.org/m/410297 ORGANIZER:upalmahbub@yahoo.com SEQUENCE:18 SUMMARY:Bridging the Gap Between Deep Learning Theory and Practice URL;VALUE=URI:https://events.vtools.ieee.org/m/410297 X-ALT-DESC:Description: <br /><p>Despite the widespread proliferation of ne ural networks\, the mechanisms through which they operate so successfully are not well understood.&nbsp\; In this talk\, we will first explore empir ical and theoretical investigations into neural network training and gener alization and what they can tell us about why deep learning works.&nbsp\; Then\, we will examine a recent line of work on algorithm learning.&nbsp\; While neural networks typically excel at pattern matching tasks\, we cons ider whether neural networks can learn algorithms that scale to problem in stances orders of magnitude larger than those seen during training.</p><br /><br />Agenda: <br /><p>- Invited talk from Micah Goldblum\, postdoctora l research fellow at New York University working with <a href="https://res earch.facebook.com/people/lecun-yann/" target="_blank" rel="external nofol low noopener">Yann LeCun</a> and <a href="https://cims.nyu.edu/~andrewgw/" target="_blank" rel="external nofollow noopener">Andrew Gordon Wilson</a> .</p>\n<p>- Q/A Session</p> END:VEVENT END:VCALENDAR