1 Cover "Hi everyone. I'm a doctor at New York People's Hospital. Today, let's chat about how reinforcement learning is making healthcare better. Hear ...

油管上看了UMN化学系某教授的工作日常有感，写完不禁感慨自己的科研效率还是不高。 1 前期准备工作 回顾之前在阅读相关文献时所做的笔记 更新相关论文的检索 确定论文的读者 确定论文的种类：研究 or 综述 or 教学 想投稿的杂志 最重要的读者是审稿人 2 整体构思 撰写初稿是最具创造 ...

1 摘要 Despite the tremendous efforts by social media platforms and factcheck services for fake news detection, fake news and misinformation still spre ...

1 Transformer 1.1 Representation Learning for NLP image.png NNs build representations of input data as vectors/embeddings, which encode useful sta ...

参考资料： 数值分析 第五版 (李庆扬，王能超，易大义) 数值计算方法与算法（张韵华，奚梅成，陈效群） 数值方法 （关治，陆金甫） Linear Algebra Done Right（Sheldon Axler） 1 基础 1.1 误差 1.1.1 误差定义 绝对误差: 设 为精确值, ...

1 📑Metadata The Transformer architecture has become a dominant choice in many domains, such as natural language processing and computer vision. Yet, ...

1 📑Metadata 摘要 - Recently, Transformer model, which has achieved great success in many artificial intelligence fields, has demonstrated its great pote ...

相关视频：李沐：如何读论文 1 阅读流程 论文通常分为6个部分：Title、Abstract、Introduction (包含Related works)、Method、Experiments、Conclusion。 沐神建议大家分三遍来读一篇论文。 第一遍，目的是大概知道论文在讲什么，适不适 ...

1 📑Metadata [!abstract] 摘要 Graph neural networks are powerful architectures for structured datasets. However, current methods struggle to represent ...

1 📑Metadata 摘要 [[In-context learning]] is the ability of a pretrained model to adapt to novel and diverse downstream tasks by conditioning on prompt ...