Density-Softmax: Efficient Test-time Model for Uncertainty Estimation and Robustness under Distribution Shifts
Fuente:
arXiv
Guardado en:
| Autores principales: | , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2023
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866913365298774016 |
|---|---|
| author | Bui, Ha Manh Liu, Anqi |
| author_facet | Bui, Ha Manh Liu, Anqi |
| contents | Sampling-based methods, e.g., Deep Ensembles and Bayesian Neural Nets have become promising approaches to improve the quality of uncertainty estimation and robust generalization. However, they suffer from a large model size and high latency at test-time, which limits the scalability needed for low-resource devices and real-time applications. To resolve these computational issues, we propose Density-Softmax, a sampling-free deterministic framework via combining a density function built on a Lipschitz-constrained feature extractor with the softmax layer. Theoretically, we show that our model is the solution of minimax uncertainty risk and is distance-aware on feature space, thus reducing the over-confidence of the standard softmax under distribution shifts. Empirically, our method enjoys competitive results with state-of-the-art techniques in terms of uncertainty and robustness, while having a lower number of model parameters and a lower latency at test-time. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2302_06495 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | Density-Softmax: Efficient Test-time Model for Uncertainty Estimation and Robustness under Distribution Shifts Bui, Ha Manh Liu, Anqi Machine Learning Sampling-based methods, e.g., Deep Ensembles and Bayesian Neural Nets have become promising approaches to improve the quality of uncertainty estimation and robust generalization. However, they suffer from a large model size and high latency at test-time, which limits the scalability needed for low-resource devices and real-time applications. To resolve these computational issues, we propose Density-Softmax, a sampling-free deterministic framework via combining a density function built on a Lipschitz-constrained feature extractor with the softmax layer. Theoretically, we show that our model is the solution of minimax uncertainty risk and is distance-aware on feature space, thus reducing the over-confidence of the standard softmax under distribution shifts. Empirically, our method enjoys competitive results with state-of-the-art techniques in terms of uncertainty and robustness, while having a lower number of model parameters and a lower latency at test-time. |
| title | Density-Softmax: Efficient Test-time Model for Uncertainty Estimation and Robustness under Distribution Shifts |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2302.06495 |