Wristband Gaussian Loss deterministically Gaussianizes point embeddings via sphere-interval decomposition with a Lean-verified proof that the pushforward is uniform iff the source is N(0,I_d), plus efficient repulsion-energy computation and application to deterministic Gaussian autoencoders.
Title resolution pending
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.LG 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
Defines the neural codebook channel K_{e→d}(j|i) and proves a Bernoulli-KL bound on encoder-decoder mismatch in VAEs that cannot be recovered from marginal histograms or mutual information.
citing papers explorer
-
The Wristband Gaussian Loss: Deterministic, Composable Latents via a Sphere-Interval Decomposition
Wristband Gaussian Loss deterministically Gaussianizes point embeddings via sphere-interval decomposition with a Lean-verified proof that the pushforward is uniform iff the source is N(0,I_d), plus efficient repulsion-energy computation and application to deterministic Gaussian autoencoders.
-
Lost and Found in Translation: Variational Diagnostics for Neural Codebook Channels
Defines the neural codebook channel K_{e→d}(j|i) and proves a Bernoulli-KL bound on encoder-decoder mismatch in VAEs that cannot be recovered from marginal histograms or mutual information.