Pareto frontier literally off the charts

• upload the run-script.ipynb file to google colab or modal
• choose A100
• Hit run all

Every DL approach on ARC today trains a supervised algorithm[1]

This is dumb.
A self-supervised compression step will obviously perform better:

• There is new information in the input grids and private puzzles that is currently uncompressed
• Test grids have distribution shifts. Compression will push these grids into distribution

Implementation details: New pareto frontier on ARC-AGI For more reasoning behind the approach, read my blog on Why all ARC solvers fail today

Performance - 27.5% on ARC-1 public eval Total Compute cost - $1.8

• ~127min on 40GB A100 for training (1.2$)
• ~49min on 80GB A100 for inference (0.6$)

This is early performance. I was too GPU poor to do hyperparameter sweeps.

I should be able to push to 35% with just basic sweeps

I expect to hit 50% with a few obvious research ideas

[1]: CompressARC is an exception, but that compresses each task individually. Mine jointly compresses all tasks together. This gives better performance at lower cost, and is more "bitter lesson" pilled.