Make Leniax simulation fully differentiable
Closed this issue · 2 comments
morgangiraud commented
TODO:
- Make leniax pipeline differentiable relative to initial conditions
- Make leniax pipeline differentiable relative to kernel parameters
- Make leniax pipeline differentiable relative to growth function parameters
- Make leniax pipeline differentiable relative to world parameters
Step 1 can be tested with a search over initialization using SGD, a RL-style objectify on the length andsome CPPN/SIREN network.
morgangiraud commented
All right, Leniax is now fully differentiable and I've already been able to learn the Lenia #0 Kernel from noise.
Let's see if we can use a neural network for the growth function now.
morgangiraud commented
Let's close this issue now. It's more interesting to replicate Neural CA up to Neural Lenia which generalize the pipeline.