Metamaterial geometry optimized by backpropagating sensing loss through differentiable simulation improves accuracy up to 5x or cuts required sensors by nearly 10x.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
q-bio.NC 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Sensing Intelligence as a Trainable Metamaterial Property
Metamaterial geometry optimized by backpropagating sensing loss through differentiable simulation improves accuracy up to 5x or cuts required sensors by nearly 10x.