A co-design of quantization, lookup-table sharing, and analog in-memory circuits lets large KAN recommendation models (39-63MB) scale with 28-41x area growth for 500K-807Kx parameter growth, with 0.11-0.23% accuracy loss in simulation.
An 89 TOPS/W and 16.3 TOPS/mm2 alldigital SRAM- based full-precision compute-in memory macro in 22 nm for machine- learning edge applications,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.AR 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Hardware Acceleration of Kolmogorov-Arnold Network (KAN) in Large-Scale Systems
A co-design of quantization, lookup-table sharing, and analog in-memory circuits lets large KAN recommendation models (39-63MB) scale with 28-41x area growth for 500K-807Kx parameter growth, with 0.11-0.23% accuracy loss in simulation.