存储引擎:对数据库进行写入和读取,不同存储引擎使用不同的格式存放数据
Mysql服务器支持多种存储引擎,默认是InnoDB
对数据进行操作需要在内存上进行,而数据是存储在磁盘上的,磁盘读写比内存读写慢很多,从磁盘上一条一条地读数据很慢,所以InnoDB将数据划分为若干个页,以页作为磁盘和内存之间交互的基本单位,InnoDB中页的大小一般为 16 KB
MapReduce and Hadoop
Big Data Analytics
Scalable and Interpretable Data Analytics
Interactive Visualization
User’s Interpretation and Control
Statistical Learning
Statistical Patterns and Models. i.e. Machine Learning
Database
Scalable Data Processing and Data Semantics Modelling. i.e. capture information behind data
Parallel Processing
Massively Parallel Computation on Modern Hardware
Spring -- IoC
IOC 容器具有依赖注入功能的容器,它可以创建对象,IOC 容器负责实例化、定位、配置应用程序中的对象及建立这些对象间的依赖。通常 new 一个实例,是由程序员控制的,而控制反转是指该工作不由程序员来做而是交给Spring容器来做。在Spring中BeanFactory是IOC容器的实际代表者。
Overview of Spring
暑假回国隔离中,想来自学一下java框架,以免实习的时候一问三不知。目前对开发的印象就是大二的时候在实验室,有mapper还有一些别的xml文件,当时就是简单的运用数据库、注册api之类的。虽然在写代码,但是没有一个总体对框架的概念。当年应该是简单的Mybatis,最近看了看决定还是从最popular的Spring开始看吧,过程可能很煎熬,毕竟想要同步看用法和源码。之前看hashmap源码的时候就在想,如果当时自学java查用法的时候顺手就把数据结构的源码看了,是不是未来会方便很多。Anyway就在blog记录一下自学Spring的进程啦,真就从0开始呗!
Reinforcement Learning
RL:
- figure out what leads to good result / bad result
- do something to get the good result
Adaptive Dynamic Programming:
- Learn the model (transition & reward function): supervised learning
- policy evaluation
Latest on Transformers -- New Techniques after BERT
Review on Transformers
self-attention
Idea: building a relationship between any 2 inputs (including itself)
RNN compute the hidden states sequentially, while in Attention they are computed in parallel.
q: query (to match others) $q^i = W^qa^i$
k: key (to be matched) $k^i = W^ka^i$
v: value (to be extracted) $v^i = W^va^i$
$\alpha_{i,j}$: the strength of the relationship between $x_i$ and $x_j$, We use each query $q$ to match each key $k$.