Training connections as well as gate/LUT functions lets deep logic-gate and lookup-table networks match fixed-wiring accuracy with far fewer gates (e.g. 8k vs ~384k on MNIST).
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Fully Trainable Deep Differentiable Logic Gate Networks and Lookup Table Networks
Training connections as well as gate/LUT functions lets deep logic-gate and lookup-table networks match fixed-wiring accuracy with far fewer gates (e.g. 8k vs ~384k on MNIST).