Resilient Practical Test-Time Adaptation: Soft Batch Normalization Alignment and Entropy-driven Memory Bank
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866917575381745664 |
|---|---|
| author | Zhou, Xingzhi Tian, Zhiliang Cheung, Ka Chun See, Simon Zhang, Nevin L. |
| author_facet | Zhou, Xingzhi Tian, Zhiliang Cheung, Ka Chun See, Simon Zhang, Nevin L. |
| contents | Test-time domain adaptation effectively adjusts the source domain model to accommodate unseen domain shifts in a target domain during inference. However, the model performance can be significantly impaired by continuous distribution changes in the target domain and non-independent and identically distributed (non-i.i.d.) test samples often encountered in practical scenarios. While existing memory bank methodologies use memory to store samples and mitigate non-i.i.d. effects, they do not inherently prevent potential model degradation. To address this issue, we propose a resilient practical test-time adaptation (ResiTTA) method focused on parameter resilience and data quality. Specifically, we develop a resilient batch normalization with estimation on normalization statistics and soft alignments to mitigate overfitting and model degradation. We use an entropy-driven memory bank that accounts for timeliness, the persistence of over-confident samples, and sample uncertainty for high-quality data in adaptation. Our framework periodically adapts the source domain model using a teacher-student model through a self-training loss on the memory samples, incorporating soft alignment losses on batch normalization. We empirically validate ResiTTA across various benchmark datasets, demonstrating state-of-the-art performance. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2401_14619 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Resilient Practical Test-Time Adaptation: Soft Batch Normalization Alignment and Entropy-driven Memory Bank Zhou, Xingzhi Tian, Zhiliang Cheung, Ka Chun See, Simon Zhang, Nevin L. Machine Learning Test-time domain adaptation effectively adjusts the source domain model to accommodate unseen domain shifts in a target domain during inference. However, the model performance can be significantly impaired by continuous distribution changes in the target domain and non-independent and identically distributed (non-i.i.d.) test samples often encountered in practical scenarios. While existing memory bank methodologies use memory to store samples and mitigate non-i.i.d. effects, they do not inherently prevent potential model degradation. To address this issue, we propose a resilient practical test-time adaptation (ResiTTA) method focused on parameter resilience and data quality. Specifically, we develop a resilient batch normalization with estimation on normalization statistics and soft alignments to mitigate overfitting and model degradation. We use an entropy-driven memory bank that accounts for timeliness, the persistence of over-confident samples, and sample uncertainty for high-quality data in adaptation. Our framework periodically adapts the source domain model using a teacher-student model through a self-training loss on the memory samples, incorporating soft alignment losses on batch normalization. We empirically validate ResiTTA across various benchmark datasets, demonstrating state-of-the-art performance. |
| title | Resilient Practical Test-Time Adaptation: Soft Batch Normalization Alignment and Entropy-driven Memory Bank |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2401.14619 |