BEGIN:VCALENDAR
VERSION:2.0
PRODID:icalendar-ruby
CALSCALE:GREGORIAN
METHOD:PUBLISH
BEGIN:VTIMEZONE
TZID:Europe/Vienna
BEGIN:DAYLIGHT
DTSTART:20230326T030000
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3
TZNAME:CEST
END:DAYLIGHT
BEGIN:STANDARD
DTSTART:20231029T020000
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10
TZNAME:CET
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260424T114228Z
UID:64a2d6ce68d9d435526149@ist.ac.at
DTSTART:20231010T160000
DTEND:20231010T170000
DESCRIPTION:Speaker: Maayan Levy\nhosted by Vogels group\nAbstract: How cha
 nges in synaptic connections lead to learning and memory is a central ques
 tion in Neuroscience. Previous modeling efforts have focused on biological
 ly realistic learning rules and dynamics\, but no known model has been sho
 wn to perform successful learning while preserving a realistic distributio
 n of synaptic weights (lognormal)\, as found experimentally. It thus remai
 ns unknown how constraining the initial and final distribution of synaptic
  weights impacts the operational principles of the learning rule as well a
 s the capacity of the network. Here we set up a spiking neural network wit
 h a "trading floor" of synaptic weights\, where weights can be swapped bet
 ween excitatory synapses according to a functional or structural plasticit
 y rule. Swapping allows for retention of the distribution while remaining 
 agnostic to the implementation of the learning rule. We then test the netw
 ork for pattern completion of corrupted visual inputs as a measure of memo
 ry. We find that while both functional and structural rules lead to patter
 n completion\, the minimum synaptic change necessary to store a pattern an
 d the resulting dynamics differ. We find that functional plasticity requir
 es broad reconfiguration of weights\, but is self-stabilizing and does not
  lead to runaway excitation. In contrast\, structural plasticity of a smal
 l number of connections is sufficient for learning\, yet results in aberra
 nt network behavior. To explore the capacity of the network\, we swap syna
 pses in response to multiple stimuli of increasing sizes. This work thus t
 ies together network topology\, capacity and limitations of synaptic plast
 icity rules.
LOCATION:Central Bldg / O1 / Mondi 2 (I01.O1.008)\, ISTA
ORGANIZER:mmosiash@ist.ac.at
SUMMARY:Maayan Levy: A trade-floor of synaptic weights explores the operati
 ng principles of plasticity in networks with preserved weight distribution
URL:https://talks-calendar.ista.ac.at/events/4492
END:VEVENT
END:VCALENDAR