esychology

5 ms·

1.
by esychology 3mo ago
In normal NCA cells are pixels and they can perceive their neighboring pixels (cells can't move). In NPA cells are particles and they can perceive all particles in a support radius around them and these particles can move freely. Does
2.
by esychology 3mo ago
Indeed! The system has good regeneration capabilities but it certainly has limits. The particles can only grow reliably if they start from the egg-like initial condition. If we switch the rules mid rollout, we would get a messed up morpholo
3.
by esychology 3mo ago
I uploaded the videos here: https://drive.google.com/drive/folders/1V8XFzq2VkZXKG7Tw8ICv...
4.
by esychology 3mo ago
I really loved the distill articles. Too bad it was not continued anymore...
5.
by esychology 3mo ago
Thanks! Yeah I think it should be possible though it requires making the cell division/splitting a differentiable operation. But nontheless, this is indeed a very interesting and promising direction to pursue.
6.
Show HN: Neural Particle Automata(selforg-npa.github.io)
88 points by esychology 3mo ago|19 comments
7.
by esychology 4mo ago
One NN per pattern/image (instance based training).
8.
by esychology 4mo ago
Checkout the Isotropic NCA blog from the Google Zurich team: https://google-research.github.io/self-organising-systems/is...
9.
by esychology 4mo ago
Thank you for the kind comment! Please reach out, I'm happy to have a chat.
10.
by esychology 4mo ago
That's not possible in the current demo but this sounds like an interesting feature to work on and add!
11.
by esychology 4mo ago
Texture sampling retrieves pixels by coordinate, while NCA grows them from local rules with no global lookup. The weights are actually ~3× smaller than JPEG-compressed texture maps, so it's not just memorizing the image either. The mor
12.
by esychology 4mo ago
This is awesome! The strength of the flow/advection is a bit too high imo. Maybe increase some viscosity parameter?
13.
by esychology 4mo ago
I think performance is not the only issue for scaling to larger grids. CUDA Convolution implementation already utilizes coalescing to improve performance. The main bottleneck is that in larger grids, cells are further apart, and it takes m
14.
by esychology 4mo ago
The input to the NN is just the 3x3 neighborhood around a cell. We can overlap two NNs on the same grid (through interpolation). Checkout https://meshnca.github.io to see the effect. When the brush is in graft mode, it basical
15.
by esychology 4mo ago
If you're familiar with CAs (e.g. Conway's Game of Life), you can think of a NeuralCA as a CA where the update rule is given by a neural network. Here we optimize the neural net weights so that it behaves a certain way (e.g. grow
16.
by esychology 4mo ago
The NeuralCA both generates and maintains the pattern. Because the NCA was not exposed to damage or erasure during training, its regeneration capability is a purely emergent phenomenon. However, this ability remains somewhat brittle, partic
17.
by esychology 4mo ago
yeah normally NCAs have a sense of up and left. There are some isotropic variants that make the perception fully rotation-invariant.
18.
by esychology 4mo ago
haha yes, also the same with the worm
19.
Show HN: High-Res Neural Cellular Automata(cells2pixels.github.io)
208 points by esychology 4mo ago|54 comments