文獻記錄－ADVERSARIAL TRAINING METHODS FOR SEMI-SUPERVISED TEXT CLASSIFICATION
文獻地址： https://arxiv.org/pdf/1605.07725.pdf 摘要： ＂對抗訓練＂提供了一種正則化的監督學習算法． ＂虛擬對抗訓練＂能夠將監督學習算法拓展到半監督學習的環境中． 兩者都是對輸入向量進行較小的擾動，不適用於稀疏高維的輸入向量． 從而提升泛化能力． 簡介： 先前的對抗訓練主要應用與圖像分類，本文中要拓展到文本分類任務和序列模型． 本文的對抗訓練就是對輸入相連進行
相關文章
- 1. 5.Adversarial Training Methods for Semi-Supervised Text Classification
- 2. Adversarial Multi-task Learning for Text Classification
- 3. 【論文筆記】Adversarial Multi-task Learning for Text Classification
- 4. Text Classification
- 5. 文獻閱讀筆記—Universal Language Model Fine-tuning for Text Classification
- 6. Generative Adversarial Networks: An Overview文獻閱讀筆記
- 7. AAAI2018-Long Text Generation via Adversarial Training with Leaked Information論文筆記
- 8. 文本挖掘-Text-mining-based Fake News Detection Using Ensemble Methods -文獻閱讀筆記
- 9. 文獻閱讀筆記electra: pre-training text encoders as discriminators rather than generators
- 10. 《Long Text Generation via Adversarial Training with Leaked Information》閱讀筆記
- 更多相關文章...
- • ADO 添加記錄 - ADO 教程
- • ADO 更新記錄 - ADO 教程
- • Tomcat學習筆記（史上最全tomcat學習筆記）
- • Scala 中文亂碼解決
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. 5.Adversarial Training Methods for Semi-Supervised Text Classification
- 2. Adversarial Multi-task Learning for Text Classification
- 3. 【論文筆記】Adversarial Multi-task Learning for Text Classification
- 4. Text Classification
- 5. 文獻閱讀筆記—Universal Language Model Fine-tuning for Text Classification
- 6. Generative Adversarial Networks: An Overview文獻閱讀筆記
- 7. AAAI2018-Long Text Generation via Adversarial Training with Leaked Information論文筆記
- 8. 文本挖掘-Text-mining-based Fake News Detection Using Ensemble Methods -文獻閱讀筆記
- 9. 文獻閱讀筆記electra: pre-training text encoders as discriminators rather than generators
- 10. 《Long Text Generation via Adversarial Training with Leaked Information》閱讀筆記