Neural quantum states via weight-space programming
A research project applying recent MIT weight-space programming techniques to neural quantum states — neural networks that represent a quantum wavefunction. Novel networks were produced for a few hundred dollars of cloud GPU, versus an estimated tens of thousands of dollars with traditional training.
What it addresses
Training neural quantum state networks in the traditional manner would probably cost tens of thousands of dollars.
Shared over time
Each entry reflects what was shared on that date. Details may have changed since then.
Neural quantum states via weight-space programming
A research project applying recent MIT weight-space programming techniques to neural quantum states — neural networks that represent a quantum wavefunction. Novel networks were produced for a few hundred dollars of cloud GPU, versus an estimated tens of thousands of dollars with traditional training.
Problem: Training neural quantum state networks in the traditional manner would probably cost tens of thousands of dollars.
Posted by @anotherpaulg · Original comment ↗
“I've been using the recent MIT weight-space programming techniques to work with neural quantum states, which are neural networks that represent a quantum wavefunction. I was able to produce a bunch of novel networks for a few hundred dollars of cloud GPU, which would have probably cost tens of thousands to train in the traditional manner.”