Relational Deep Reinforcement Learning
Abstract 我們介紹了一種深度強化學習的方法，它通過結構化感知和關係推理提高了傳統方法的效率、泛化能力和可解釋性。它使用self-attention來迭代推理場景中實體之間的關係，並指導無模型策略。我們的結果表明，在一種名爲Box-World的新的導航和規劃任務中，我們的agent找到了可解釋的解決方案，這些解決方案在樣本複雜性、泛化到比訓練期間所經歷的更復雜的場景的能力以及整體性能方面都有
相關文章
- 1. Reinforcement learning and Deep learning
- 2. Deep Reinforcement Learning
- 3. Generating Text with Deep Reinforcement Learning
- 4. Continuous control with Deep Reinforcement Learning
- 5. Deep Reinforcement Learning: Pong from Pixels
- 6. Playing Atari with Deep Reinforcement Learning
- 7. Deep Reinforcement Learning for Dialogue Generation
- 8. Deep Reinforcement Learning for Join Order Enumeration
- 9. Playing atari with deep reinforcement learning
- 10. Deep Reinforcement Learning amidst Lifelong Non-Stationarity
- 更多相關文章...
- • XQuery 添加元素 和屬性 - XQuery 教程
- • XQuery FLWOR 表達式 - XQuery 教程
- • Java Agent入門實戰（一）-Instrumentation介紹與使用
- • Java Agent入門實戰（三）-JVM Attach原理與使用
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。
最新文章
歡迎關注本站公眾號,獲取更多信息
相關文章
- 1. Reinforcement learning and Deep learning
- 2. Deep Reinforcement Learning
- 3. Generating Text with Deep Reinforcement Learning
- 4. Continuous control with Deep Reinforcement Learning
- 5. Deep Reinforcement Learning: Pong from Pixels
- 6. Playing Atari with Deep Reinforcement Learning
- 7. Deep Reinforcement Learning for Dialogue Generation
- 8. Deep Reinforcement Learning for Join Order Enumeration
- 9. Playing atari with deep reinforcement learning
- 10. Deep Reinforcement Learning amidst Lifelong Non-Stationarity