Adversarially Learned One-Class Classifier for Novelty Detection
CVPR-2018 Novelty detection 是識別在某些方面與訓練觀察（目標類別）不一樣的觀察的過程。實際上，Novelty class 在訓練期間一般不存在，採樣不良或定義不明確。所以，一類分類器能夠有效地模擬這些問題。然而，因爲來自 Novelty class 的數據不可用，訓練端到端深度網絡是一項繁瑣的任務。在本文中，受到生成對抗網絡在無監督和半監督環境中訓練深度模型的成功的啓發
相關文章
- 1. 論文速讀:Adversarially Learned One-Class Classifier for Novelty Detection [CVPR 2018]
- 2. Adversarially Learned One-Class Classifier for Novelty Detection（用於單類異常檢測的對抗學習分類器）
- 3. Kernel PCA for Novelty Detection
- 4. 由Adversarially Learned Inference引發
- 5. A LEARNED REPRESENTATION FOR ARTISTIC STYLE
- 6. Towards Adversarially Robust Object Detection 論文筆記
- 7. Overcoming Classifier Imbalance for Long-tail Object Detection with Balanced Group Softmax(CVPR20)
- 8. 年齡及性別預測（2）AgeNet: Deeply Learned Regressor and Classifier for Robust Apparent Age Estimation
- 9. Maximum Classifier Discrepancy for Unsupervised Domain Adaptation
- 10. Unprocessing Images for Learned Raw Denoising
- 更多相關文章...
- • Swift for 循環 - Swift 教程
- • Scala for循環 - Scala教程
- • PHP開發工具
- • C# 中 foreach 遍歷的用法
相關標籤/搜索
每日一句
-
每一个你不满意的现在，都有一个你没有努力的曾经。
歡迎關注本站公眾號,獲取更多信息
相關文章
- 1. 論文速讀:Adversarially Learned One-Class Classifier for Novelty Detection [CVPR 2018]
- 2. Adversarially Learned One-Class Classifier for Novelty Detection（用於單類異常檢測的對抗學習分類器）
- 3. Kernel PCA for Novelty Detection
- 4. 由Adversarially Learned Inference引發
- 5. A LEARNED REPRESENTATION FOR ARTISTIC STYLE
- 6. Towards Adversarially Robust Object Detection 論文筆記
- 7. Overcoming Classifier Imbalance for Long-tail Object Detection with Balanced Group Softmax(CVPR20)
- 8. 年齡及性別預測（2）AgeNet: Deeply Learned Regressor and Classifier for Robust Apparent Age Estimation
- 9. Maximum Classifier Discrepancy for Unsupervised Domain Adaptation
- 10. Unprocessing Images for Learned Raw Denoising