Viewing profile — pleshkov

It's close but not the same. Kernel PCA lifts all D coordinates which gives M around 525k at D = 1024. In the post I do PCA first to reduce D to d = 256, then lift only those d coo…

I agree that we don't want to reconstruct the whole vector while retrieval and it makes poly-AE toy-like at the current state non production ready. My main interest here in the jus…

Just checked the normalization point. You were partially right, sqrt-normalization makes the difference x2 less. I'm updating the numbers in the post. Interesting moment. I did a s…

The polynomial lift in this post originally came out of an unsuccessful experiment with hyperbolic embeddings. The idea was to embed corpora into a hyperbolic ball (anisotropic emb…

Fair point — lam is technically a hyperparameter. In practice I used lam=1e-3 (the default in the code) across all four models without tuning, and the gap to PCA is robust enough t…

Good catch, this is the obvious ablation I should have included. I'll re-run with per-axis normalized PCA as a separate baseline and post numbers in this thread tomorrow. Prior: I …

Author here. Fair characterization, and a fair critique on the geometric story. A few clarifications. I don't claim {x_i, x_i·x_j} is the right lift specifically — the post itself …

Author here — questions and pushback both welcome.

[dead]