<?xml version="1.0" encoding="utf-8" ?><rss version="2.0"><channel><title><![CDATA[taurusong的博客]]></title><description><![CDATA[]]></description><link>https://blog.youkuaiyun.com/taurusong</link><language>zh-cn</language><generator>https://blog.youkuaiyun.com/</generator><copyright><![CDATA[Copyright &copy; taurusong]]></copyright><item><title><![CDATA[读文章：Few-Shot Unsupervised Image-to-Image Translation]]></title><link>https://blog.youkuaiyun.com/taurusong/article/details/129264413</link><guid>https://blog.youkuaiyun.com/taurusong/article/details/129264413</guid><author>taurusong</author><pubDate>Thu, 02 Mar 2023 15:46:31 +0800</pubDate><description><![CDATA[Humans are remarkably good at generalization.人类擅长演绎。This problem is inherently ill-posed as it attempts to recover the joint distribution using samples from marginal distributions非配对图像到图像的翻译是从边缘分布的例子推测联合分布，因此是ill-posed.]]></description><category></category></item><item><title><![CDATA[读文章：ChatGPT is not all you need. A State of the Art Review of large Generative AI models]]></title><link>https://blog.youkuaiyun.com/taurusong/article/details/129256293</link><guid>https://blog.youkuaiyun.com/taurusong/article/details/129256293</guid><author>taurusong</author><pubDate>Tue, 28 Feb 2023 15:38:43 +0800</pubDate><description><![CDATA[ChatGPT 和 Stable Diffusion 的出现，掀动了一股生成式AI热潮。生成式AI能够有效的和创造性地做事情。]]></description><category></category></item><item><title><![CDATA[读文章：Revisiting Bellman Errors for Offline Model Selection]]></title><link>https://blog.youkuaiyun.com/taurusong/article/details/128896092</link><guid>https://blog.youkuaiyun.com/taurusong/article/details/128896092</guid><author>taurusong</author><pubDate>Mon, 06 Feb 2023 12:27:17 +0800</pubDate><description><![CDATA[*Offline reinforcement learning (ORL)离线强化学习：**与在线强化学习相对的一个概念，ORL有固定的数据集，数据是从静态数据集获取的；在线强化学习与环境有交互演进，数据是动态的从环境获取的。**优点：**由于不需要和环境实时交互，ORL为研究的经济性和安全性提供了保障，在自动驾驶、健康医疗等领域中可以使用。]]></description><category></category></item></channel></rss>