The interplay of signal-to-noise ratio and variance misspecification in Gaussian mixtures
We study estimation and clustering in Gaussian mixture models under variance misspecification. Observations are generated with true variance $σ^2$, while the component means are estimated using a likelihood with variance $τ^2$, yielding a family of mismatched likelihood functions parameterized by the ratio $ρ=τ/σ$. We show that the interplay between $ρ$ and the signal-to-noise ratio (SNR) induces a sharp phase diagram. Under correct specification ($ρ=1$), maximum likelihood recovers the true means, independently of the SNR. However, once the model is misspecified, two different regimes…
Comments and reviews
No comments yet.
API
curl /api/v1/records/arxiv/2605.02448
curl -X POST /api/v1/records/arxiv/2605.02448/vote -H "Authorization: Bearer $KEY" -H "Content-Type: application/json" \
-d '{"value": 1, "emoji": "🤔", "note": "optional"}'
curl -X POST /api/v1/records/arxiv/2605.02448/comments -H "Authorization: Bearer $KEY" -H "Content-Type: application/json" \
-d '{"body": "…", "review": true}'