Viewing profile — esychology

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…

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 swi…

I uploaded the videos here: https://drive.google.com/drive/folders/1V8XFzq2VkZXKG7Tw8ICv...

I really loved the distill articles. Too bad it was not continued anymore...

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 p…

Neural CAs model self-organizing pattern formation on grids. Now the grid is gone. Each cell is an agentic particle that can move freely in space and change its state. While each p…

One NN per pattern/image (instance based training).

Checkout the Isotropic NCA blog from the Google Zurich team: https://google-research.github.io/self-organising-systems/is...

Thank you for the kind comment! Please reach out, I'm happy to have a chat.

That's not possible in the current demo but this sounds like an interesting feature to work on and add!

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…

This is awesome! The strength of the flow/advection is a bit too high imo. Maybe increase some viscosity parameter?

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 th…

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 effec…

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 we…

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 phenom…

yeah normally NCAs have a sense of up and left. There are some isotropic variants that make the perception fully rotation-invariant.

haha yes, also the same with the worm

Neural CAs model self-organizing pattern formation. Now they can generate patterns at HD resolution in real-time, enabled by turning each CA cell into a Neural Field. Try 3 demos: …