BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//cfp.riscv-europe.org//eu-summit-2026//speaker//U3BGSJ
BEGIN:VTIMEZONE
TZID:CET
BEGIN:STANDARD
DTSTART:20001029T040000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10
TZNAME:CET
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20000326T030000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3
TZNAME:CEST
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:pretalx-eu-summit-2026-QM3SVB@cfp.riscv-europe.org
DTSTART;TZID=CET:20260611T132000
DTEND;TZID=CET:20260611T133000
DESCRIPTION:This work proposes an open-source framework that leverages both
  the Zfh (scalar float16) and the Zvfh (vector float16) extensions to enab
 le complete on-device training on resource-constrained RISC-V single-core.
  On top of reducing the memory footprint by about 50% as compared to using
  float32\, our approach facilitates transfer learning and fine-tuning scen
 arios by incorporating layer-freezing capabilities. Our work builds onto A
 IfES an open-source\, modular and generic DNN training and inference frame
 work for embedded systems that can be extended with custom hardware-specif
 ic functions.
DTSTAMP:20260522T162433Z
LOCATION:Poster Island C
SUMMARY:An Open-Source Framework to Enable Float16 On-Device Training on RI
 SC-V Single-Core - Benjamin Hubinet
URL:https://cfp.riscv-europe.org/eu-summit-2026/talk/QM3SVB/
END:VEVENT
END:VCALENDAR