Expand description
Mutation operators for real-valued genomes.
Every mutation is an in-place-ish tensor transform: it consumes a
population tensor and returns a new one with noise added per gene.
Tensor ops already compose nicely with Burn’s JIT fuser, so these
functions are just thin wrappers around randn_like-style patterns.
Functions§
- bit_
flip_ mutation - Bit-flip mutation on a binary
Tensor<B, 2, Int>population. - gaussian_
mutation - Isotropic Gaussian mutation with scalar σ.
- gaussian_
mutation_ per_ row - Per-row anisotropic Gaussian mutation.
- uniform_
reset - Uniform-reset mutation with per-gene probability
p.