On the Shape of Latent Variables in a Denoising VAE-MoG: A Posterior Sampling-Based Study
Fuente:
arXiv
Guardado en:
| Autor principal: | |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866911184089776128 |
|---|---|
| author | Bascuñán, Fernanda Zapata |
| author_facet | Bascuñán, Fernanda Zapata |
| contents | In this work, we explore the latent space of a denoising variational autoencoder with a mixture-of-Gaussians prior (VAE-MoG), trained on gravitational wave data from event GW150914. To evaluate how well the model captures the underlying structure, we use Hamiltonian Monte Carlo (HMC) to draw posterior samples conditioned on clean inputs, and compare them to the encoder's outputs from noisy data. Although the model reconstructs signals accurately, statistical comparisons reveal a clear mismatch in the latent space. This shows that strong denoising performance doesn't necessarily mean the latent representations are reliable highlighting the importance of using posterior-based validation when evaluating generative models. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_25382 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | On the Shape of Latent Variables in a Denoising VAE-MoG: A Posterior Sampling-Based Study Bascuñán, Fernanda Zapata Machine Learning Hardware Architecture In this work, we explore the latent space of a denoising variational autoencoder with a mixture-of-Gaussians prior (VAE-MoG), trained on gravitational wave data from event GW150914. To evaluate how well the model captures the underlying structure, we use Hamiltonian Monte Carlo (HMC) to draw posterior samples conditioned on clean inputs, and compare them to the encoder's outputs from noisy data. Although the model reconstructs signals accurately, statistical comparisons reveal a clear mismatch in the latent space. This shows that strong denoising performance doesn't necessarily mean the latent representations are reliable highlighting the importance of using posterior-based validation when evaluating generative models. |
| title | On the Shape of Latent Variables in a Denoising VAE-MoG: A Posterior Sampling-Based Study |
| topic | Machine Learning Hardware Architecture |
| url | https://arxiv.org/abs/2509.25382 |