Answering from Sure to Uncertain: Uncertainty-Aware Curriculum Learning for Video Question Answering
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866917184042696704 |
|---|---|
| author | Li, Haopeng Bennamoun, Mohammed Liu, Jun Rahmani, Hossein Ke, Qiuhong |
| author_facet | Li, Haopeng Bennamoun, Mohammed Liu, Jun Rahmani, Hossein Ke, Qiuhong |
| contents | While significant advancements have been made in video question answering (VideoQA), the potential benefits of enhancing model generalization through tailored difficulty scheduling have been largely overlooked in existing research. This paper seeks to bridge that gap by incorporating VideoQA into a curriculum learning (CL) framework that progressively trains models from simpler to more complex data. Recognizing that conventional self-paced CL methods rely on training loss for difficulty measurement, which might not accurately reflect the intricacies of video-question pairs, we introduce the concept of uncertainty-aware CL. Here, uncertainty serves as the guiding principle for dynamically adjusting the difficulty. Furthermore, we address the challenge posed by uncertainty by presenting a probabilistic modeling approach for VideoQA. Specifically, we conceptualize VideoQA as a stochastic computation graph, where the hidden representations are treated as stochastic variables. This yields two distinct types of uncertainty: one related to the inherent uncertainty in the data and another pertaining to the model's confidence. In practice, we seamlessly integrate the VideoQA model into our framework and conduct comprehensive experiments. The findings affirm that our approach not only achieves enhanced performance but also effectively quantifies uncertainty in the context of VideoQA. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2401_01510 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Answering from Sure to Uncertain: Uncertainty-Aware Curriculum Learning for Video Question Answering Li, Haopeng Bennamoun, Mohammed Liu, Jun Rahmani, Hossein Ke, Qiuhong Computer Vision and Pattern Recognition While significant advancements have been made in video question answering (VideoQA), the potential benefits of enhancing model generalization through tailored difficulty scheduling have been largely overlooked in existing research. This paper seeks to bridge that gap by incorporating VideoQA into a curriculum learning (CL) framework that progressively trains models from simpler to more complex data. Recognizing that conventional self-paced CL methods rely on training loss for difficulty measurement, which might not accurately reflect the intricacies of video-question pairs, we introduce the concept of uncertainty-aware CL. Here, uncertainty serves as the guiding principle for dynamically adjusting the difficulty. Furthermore, we address the challenge posed by uncertainty by presenting a probabilistic modeling approach for VideoQA. Specifically, we conceptualize VideoQA as a stochastic computation graph, where the hidden representations are treated as stochastic variables. This yields two distinct types of uncertainty: one related to the inherent uncertainty in the data and another pertaining to the model's confidence. In practice, we seamlessly integrate the VideoQA model into our framework and conduct comprehensive experiments. The findings affirm that our approach not only achieves enhanced performance but also effectively quantifies uncertainty in the context of VideoQA. |
| title | Answering from Sure to Uncertain: Uncertainty-Aware Curriculum Learning for Video Question Answering |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2401.01510 |