Bayesian adaptive learning to latent variables via Variational Bayes and Maximum a Posteriori
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866916105382002688 |
|---|---|
| author | Hu, Hu Siniscalchi, Sabato Marco Lee, Chin-Hui |
| author_facet | Hu, Hu Siniscalchi, Sabato Marco Lee, Chin-Hui |
| contents | In this work, we aim to establish a Bayesian adaptive learning framework by focusing on estimating latent variables in deep neural network (DNN) models. Latent variables indeed encode both transferable distributional information and structural relationships. Thus the distributions of the source latent variables (prior) can be combined with the knowledge learned from the target data (likelihood) to yield the distributions of the target latent variables (posterior) with the goal of addressing acoustic mismatches between training and testing conditions. The prior knowledge transfer is accomplished through Variational Bayes (VB). In addition, we also investigate Maximum a Posteriori (MAP) based Bayesian adaptation. Experimental results on device adaptation in acoustic scene classification show that our proposed approaches can obtain good improvements on target devices, and consistently outperforms other cut-edging algorithms. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2401_13766 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Bayesian adaptive learning to latent variables via Variational Bayes and Maximum a Posteriori Hu, Hu Siniscalchi, Sabato Marco Lee, Chin-Hui Audio and Speech Processing Sound In this work, we aim to establish a Bayesian adaptive learning framework by focusing on estimating latent variables in deep neural network (DNN) models. Latent variables indeed encode both transferable distributional information and structural relationships. Thus the distributions of the source latent variables (prior) can be combined with the knowledge learned from the target data (likelihood) to yield the distributions of the target latent variables (posterior) with the goal of addressing acoustic mismatches between training and testing conditions. The prior knowledge transfer is accomplished through Variational Bayes (VB). In addition, we also investigate Maximum a Posteriori (MAP) based Bayesian adaptation. Experimental results on device adaptation in acoustic scene classification show that our proposed approaches can obtain good improvements on target devices, and consistently outperforms other cut-edging algorithms. |
| title | Bayesian adaptive learning to latent variables via Variational Bayes and Maximum a Posteriori |
| topic | Audio and Speech Processing Sound |
| url | https://arxiv.org/abs/2401.13766 |