رسالة جامعية
Closing the Gap Between Classification and Retrieval Models
| العنوان: | Closing the Gap Between Classification and Retrieval Models |
|---|---|
| المؤلفون: | Taha, Ahmed |
| المساهمون: | Davis, Larry, Shrivastava, Abhinav, Digital Repository at the University of Maryland, University of Maryland (College Park, Md.), Computer Science |
| سنة النشر: | 2021 |
| المجموعة: | University of Maryland: Digital Repository (DRUM) |
| مصطلحات موضوعية: | Computer science, Computer vision, Deep learning, Feature embedding, Machine learning, Retrieval networks |
| الوصف: | Retrieval networks learn a feature embedding where similar samples are close together, and different samples are far apart. This feature embedding is essential for computer vision applications such as face/person recognition, zero-shot learn- ing, and image retrieval. Despite these important applications, retrieval networks are less popular compared to classification networks due to multiple reasons: (1) The cross-entropy loss – used with classification networks – is stabler and converges faster compared to metric learning losses – used with retrieval networks. (2) The cross-entropy loss has a huge toolbox of utilities and extensions. For instance, both AdaCos and self-knowledge distillation have been proposed to tackle low sample complexity in classification networks; also, both CAM and Grad-CAM have been proposed to visualize attention in classification networks. To promote retrieval networks, it is important to equip them with an equally powerful toolbox. Accordingly, we propose an evolution-inspired approach to tackle low sample complexity in feature embedding. Then, we propose SVMax to regularize the feature embedding and avoid model collapse. Furthermore, we propose L2-CAF to visualize attention in retrieval networks. To tackle low sample complexity, we propose an evolution-inspired training approach to boost performance on relatively small datasets. The knowledge evolution (KE) approach splits a deep network into two hypotheses: the fit-hypothesis and the reset-hypothesis. We iteratively evolve the knowledge inside the fit-hypothesis by perturbing the reset-hypothesis for multiple generations. This approach not only boosts performance but also learns a slim (pruned) network with a smaller inference cost. KE reduces both overfitting and the burden for data collection. To regularize the feature embedding and avoid model collapse, We propose singular value maximization (SVMax) to promote a uniform feature embedding. Our formulation mitigates model collapse and enables larger learning rates. SV- Max is ... |
| نوع الوثيقة: | doctoral or postdoctoral thesis |
| وصف الملف: | application/pdf |
| اللغة: | English |
| العلاقة: | https://doi.org/10.13016/nul5-lq6aTest; http://hdl.handle.net/1903/27316Test |
| DOI: | 10.13016/nul5-lq6a |
| الإتاحة: | https://doi.org/10.13016/nul5-lq6aTest http://hdl.handle.net/1903/27316Test |
| رقم الانضمام: | edsbas.BC6AE1C8 |
| قاعدة البيانات: | BASE |
| DOI: | 10.13016/nul5-lq6a |
|---|