BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Chicago
X-LIC-LOCATION:America/Chicago
BEGIN:DAYLIGHT
TZOFFSETFROM:-0600
TZOFFSETTO:-0500
TZNAME:CDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0500
TZOFFSETTO:-0600
TZNAME:CST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20230124T171524Z
LOCATION:C1-2-3
DTSTART;TZID=America/Chicago:20221117T083000
DTEND;TZID=America/Chicago:20221117T170000
UID:submissions.supercomputing.org_SC22_sess275_rpost176@linklings.com
SUMMARY:CLIP-ACQUA: CLIP Autoencoder-Based Classic-Quantum Latent Space Re
 duction
DESCRIPTION:Posters, Research Posters\n\nCLIP-ACQUA: CLIP Autoencoder-Base
 d Classic-Quantum Latent Space Reduction\n\nRivas, Zhao\n\nApplications of
  quantum machine learning algorithms are currently still being studied. Re
 cent work suggests that classical gradient descent techniques can effectiv
 ely train variational quantum circuits. We propose to train quantum variat
 ional circuits to find smaller text and image embeddings that preserve con
 trastive-learning distances based on CLIP large embeddings. This is a crit
 ical task since fine-tuning CLIP to produce low-dimensional embeddings is 
 prohibitively expensive. We introduce CLIP-ACQUA, a model trained in a sel
 f-supervised configuration from CLIP embeddings to reduce the latent space
 . We use CLIP-ACQUA on a sizeable unlabelled corpus of text and images to 
 demonstrate its effectiveness. Our experiments show that we can obtain sma
 ller latent spaces that preserve the original embedding distances inferred
  during contrastive learning. Furthermore, using our model requires no fin
 e-tuning of CLIP, preserving its original robustness and structure. The da
 ta used as a demonstration aids in modeling consumer-to-consumer online ma
 rketplaces to detect illicit activities.\n\nRegistration Category: Tech Pr
 ogram Reg Pass, Exhibits Reg Pass
END:VEVENT
END:VCALENDAR