A hybrid CGP scheme with a transformer mutation operator evolves approximate multipliers that achieve better error-power trade-offs than the EvoApproxLib library for several target constraints.
Approximating complex arithmetic circuits with formal error guaran- tees: 32-bit multipliers accomplished,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.NE 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Genetic Programming with Transformer-Based Mutation for Approximate Circuit Design
A hybrid CGP scheme with a transformer mutation operator evolves approximate multipliers that achieve better error-power trade-offs than the EvoApproxLib library for several target constraints.