google-research / self-organising-systems

Google Research

This repository contains code released by Google Research.

All datasets in this repository are released under the CC BY 4.0 International license, which can be found here: https://creativecommons.org/licenses/by/4.0/legalcode. All source files in this repository are released under the Apache 2.0 license, the text of which can be found in the LICENSE file.

---

Because the repo is large, we recommend you download only the subdirectory of interest:

• Use GitHub editor to open the project. To open the editor change the url from github.com to github.dev in the address bar.
• In the left navigation panel, right-click on the folder of interest and select download.

If you'd like to submit a pull request, you'll need to clone the repository; we recommend making a shallow clone (without history).

git clone [email protected]:google-research/google-research.git --depth=1

---

Disclaimer: This is not an official Google product.

Updated in 2023.

self-organising-systems's Issues

Exporting to Shadertoy

Hello! Thanks for sharing this work.

I noticed some cells related to exporting the network to shadertoy format, at the bottom of the newly committed μNCA notebooks.

Is there some boilerplate code that would need to surround this vec4 update function?

Fine-tuning the body during its life

Hello! I admire your work and see great potential for research and future technologies in it.

In your paper "Transformers learn in-context by gradient descent" you show that transformers have a semblance of gradient descent in a forward pass.
As you can see, is it possible to represent some of the body's self-regulation processes as an autoregressive model, such as a transformer?

And if so, does the organism exhibit its adaptation similar to gradient descent by the error of simple autoregression?

Can not run the code Transformers learn in-context by gradient descent/constructed_token_setup.ipynb

The error as following

No such file or directory: 'model_npys.zip'

When trying to run texture_nca_tf2.ipynb in Colab I got the following error:

---------------------------------------------------------------------------
FileNotFoundError                         Traceback (most recent call last)
<ipython-input-5-649be2f889ba> in <module>()
      1 #@title ⬅ run this cell to explore pretrained models! {vertical-output: true}
      2 get_ipython().system('wget -qnc --show-progress https://storage.googleapis.com/selforg/texture_ca/model_npys.zip')
----> 3 pretrained_models = np.load('model_npys.zip',  allow_pickle=True)
      4 
      5 model_groups = {}

/usr/local/lib/python3.7/dist-packages/numpy/lib/npyio.py in load(file, mmap_mode, allow_pickle, fix_imports, encoding)
    414             own_fid = False
    415         else:
--> 416             fid = stack.enter_context(open(os_fspath(file), "rb"))
    417             own_fid = True
    418 

FileNotFoundError: [Errno 2] No such file or directory: 'model_npys.zip'

The file this cell trying to download seems missing.