Hyperparameters for 5.1 (synthetic anti-causal regression experiment)

[mattyshen]

I am looking to recreate your results in Table 2 from the BIRM paper (synthetic anti-causal regression with x1, x2, and y), and am reaching out to inquire about the specific hyperparameters you used for BIRM, and all the other invariant methods if possible. I am having a hard time finding this experiment (5.1) in your released github.

I reached out to Yong Lin (at the princeton address) over email, but it seems to be deactivated.