Saved in:
| Main Author: | |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2602.23315 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915820564643840 |
|---|---|
| author | Hu, Sha |
| author_facet | Hu, Sha |
| contents | An artificial intelligence (AI) model can be viewed as a function that maps inputs to outputs in high-dimensional spaces. Once designed and well trained, the AI model is applied for inference. However, even optimized AI models can produce inference errors due to aleatoric and epistemic uncertainties. Interestingly, we observed that when inferring multiple samples based on invariant transformations of an input, inference errors can show partial independences due to epistemic uncertainty. Leveraging this insight, we propose a "resampling" based inferencing that applies to a trained AI model with multiple transformed versions of an input, and aggregates inference outputs to a more accurate result. This approach has the potential to improve inference accuracy and offers a strategy for balancing model size and performance. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_23315 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Invariant Transformation and Resampling based Epistemic-Uncertainty Reduction Hu, Sha Artificial Intelligence An artificial intelligence (AI) model can be viewed as a function that maps inputs to outputs in high-dimensional spaces. Once designed and well trained, the AI model is applied for inference. However, even optimized AI models can produce inference errors due to aleatoric and epistemic uncertainties. Interestingly, we observed that when inferring multiple samples based on invariant transformations of an input, inference errors can show partial independences due to epistemic uncertainty. Leveraging this insight, we propose a "resampling" based inferencing that applies to a trained AI model with multiple transformed versions of an input, and aggregates inference outputs to a more accurate result. This approach has the potential to improve inference accuracy and offers a strategy for balancing model size and performance. |
| title | Invariant Transformation and Resampling based Epistemic-Uncertainty Reduction |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2602.23315 |