BEGIN:VCALENDAR VERSION:2.0 PRODID:IEEE vTools.Events//EN CALSCALE:GREGORIAN BEGIN:VTIMEZONE TZID:Australia/Melbourne BEGIN:DAYLIGHT DTSTART:20261004T030000 TZOFFSETFROM:+1000 TZOFFSETTO:+1100 RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=10 TZNAME:AEDT END:DAYLIGHT BEGIN:STANDARD DTSTART:20260405T020000 TZOFFSETFROM:+1100 TZOFFSETTO:+1000 RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=4 TZNAME:AEST END:STANDARD END:VTIMEZONE BEGIN:VEVENT DTSTAMP:20260429T051629Z UID:0F6D0732-AA1A-441D-BBF2-85D598FB8117 DTSTART;TZID=Australia/Melbourne:20260429T140000 DTEND;TZID=Australia/Melbourne:20260429T150000 DESCRIPTION:Abstract:\n\nBackpropagation has powered the modern era of comp utational intelligence\, enabling breakthroughs in perception\, language\, control\, and autonomous systems. Yet as intelligent systems move into dy namic\, real-world environments\, new demands emerge: continual adaptation \, robustness under uncertainty\, energy efficiency\, and scalable autonom y. These challenges invite a deeper question — are our learning algorith ms fundamentally aligned with how intelligence itself operates?\n\nThis le cture explores predictive coding as a compelling\, brain-inspired alternat ive for credit assignment in deep systems. Rather than relying on staged f orward and backward passes with global error transport\, predictive coding formulates learning as the continuous minimization of hierarchical predic tion errors through local\, parallel\, and bidirectional interactions. Rec ent theoretical advances demonstrate that such dynamics can approximate gr adient-based optimization\, offering a principled bridge between neuroscie nce and modern machine learning.\n\nThis perspective reframes learning as an energy-minimizing dynamical process\, opening new directions in distrib uted credit assignment\, continual learning\, robust inference\, and neuro morphic implementation. By revisiting the principles of biological intelli gence\, this lecture argues that the next generation of computational inte lligence systems may emerge not from scaling existing algorithms\, but fro m rethinking the foundations of learning itself.\n\nCo-sponsored by: IEEE VIC CIS Chapter\; IEEE VIC Section\n\nSpeaker(s): Narayan Srinivasa\, \n\n Virtual: https://events.vtools.ieee.org/m/554620 LOCATION:Virtual: https://events.vtools.ieee.org/m/554620 ORGANIZER:malka_nisha@ieee.org SEQUENCE:46 SUMMARY:Rethinking Learning: Beyond Backpropagation Toward Brain-Inspired C omputational Intelligence URL;VALUE=URI:https://events.vtools.ieee.org/m/554620 X-ALT-DESC:Description: <br /><p dir="ltr">&nbsp\;</p>\n<p dir="ltr"><img s rc="https://events.vtools.ieee.org/vtools_ui/media/display/011a8392-8438-4 59f-a515-684eec256d9e"></p>\n<p dir="ltr">&nbsp\;</p>\n<p dir="ltr"><stron g>Abstract:&nbsp\;</strong></p>\n<p dir="ltr">Backpropagation has powered the modern era of computational intelligence\, enabling breakthroughs in p erception\, language\, control\, and autonomous systems. Yet as intelligen t systems move into dynamic\, real-world environments\, new demands emerge : continual adaptation\, robustness under uncertainty\, energy efficiency\ , and scalable autonomy. These challenges invite a deeper question &mdash\ ; are our learning algorithms fundamentally aligned with how intelligence itself operates?</p>\n<p dir="ltr">This lecture explores predictive coding as a compelling\, brain-inspired alternative for credit assignment in dee p systems. Rather than relying on staged forward and backward passes with global error transport\, predictive coding formulates learning as the cont inuous minimization of hierarchical prediction errors through local\, para llel\, and bidirectional interactions. Recent theoretical advances demonst rate that such dynamics can approximate gradient-based optimization\, offe ring a principled bridge between neuroscience and modern machine learning. </p>\n<p dir="ltr">This perspective reframes learning as an energy-minimiz ing dynamical process\, opening new directions in distributed credit assig nment\, continual learning\, robust inference\, and neuromorphic implement ation. By revisiting the principles of biological intelligence\, this lect ure argues that the next generation of computational intelligence systems may emerge not from scaling existing algorithms\, but from rethinking the foundations of learning itself.</p>\n<p>&nbsp\;</p> END:VEVENT END:VCALENDAR