DANCE: Doubly Adaptive Neighborhood Conformal Estimation
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2026
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866912923050311680 |
|---|---|
| author | Feng, Brandon R. Reich, Brian J. Beaglehole, Daniel Luo, Xihaier Park, David Keetae Yoo, Shinjae Huang, Zhechao Mao, Xueyu Boz, Olcay Kim, Jungeum |
| author_facet | Feng, Brandon R. Reich, Brian J. Beaglehole, Daniel Luo, Xihaier Park, David Keetae Yoo, Shinjae Huang, Zhechao Mao, Xueyu Boz, Olcay Kim, Jungeum |
| contents | The recent developments of complex deep learning models have led to unprecedented ability to accurately predict across multiple data representation types. Conformal prediction for uncertainty quantification of these models has risen in popularity, providing adaptive, statistically-valid prediction sets. For classification tasks, conformal methods have typically focused on utilizing logit scores. For pre-trained models, however, this can result in inefficient, overly conservative set sizes when not calibrated towards the target task. We propose DANCE, a doubly locally adaptive nearest-neighbor based conformal algorithm combining two novel nonconformity scores directly using the data's embedded representation. DANCE first fits a task-adaptive kernel regression model from the embedding layer before using the learned kernel space to produce the final prediction sets for uncertainty quantification. We test against state-of-the-art local, task-adapted and zero-shot conformal baselines, demonstrating DANCE's superior blend of set size efficiency and robustness across various datasets. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_20652 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | DANCE: Doubly Adaptive Neighborhood Conformal Estimation Feng, Brandon R. Reich, Brian J. Beaglehole, Daniel Luo, Xihaier Park, David Keetae Yoo, Shinjae Huang, Zhechao Mao, Xueyu Boz, Olcay Kim, Jungeum Machine Learning The recent developments of complex deep learning models have led to unprecedented ability to accurately predict across multiple data representation types. Conformal prediction for uncertainty quantification of these models has risen in popularity, providing adaptive, statistically-valid prediction sets. For classification tasks, conformal methods have typically focused on utilizing logit scores. For pre-trained models, however, this can result in inefficient, overly conservative set sizes when not calibrated towards the target task. We propose DANCE, a doubly locally adaptive nearest-neighbor based conformal algorithm combining two novel nonconformity scores directly using the data's embedded representation. DANCE first fits a task-adaptive kernel regression model from the embedding layer before using the learned kernel space to produce the final prediction sets for uncertainty quantification. We test against state-of-the-art local, task-adapted and zero-shot conformal baselines, demonstrating DANCE's superior blend of set size efficiency and robustness across various datasets. |
| title | DANCE: Doubly Adaptive Neighborhood Conformal Estimation |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2602.20652 |