Adrian Blog | Data Enthusiast

Statistics Concepts Review

"97.3% of all statistics are made-up"

“Give more than you planned to” Principal Concepts Must be able to explain concretely and concisely P value, significance level, confidence level, confidence interval Common Distirbutio...

Posted by dliu on August 15, 2021

Comprehensive Reviews of Data Scientist Interview

"Find your leakage and patch it up"

Today, let’s do a higher-level comprehensive review of what a Data Scientist must know before jumping to an interview. I understand the job market is yet thoroughly opened, but get well-prepared f...

Posted by dliu on June 5, 2021

Happy Children's Day

"Hope we all live in a peaceful world"

Read news, from various sources, various languages. Feel sad about what the person/victim suffered. Feel sad about what is happening in US, Canada. Feel sad about what we are going through in 20...

Posted by dliu on June 1, 2020

My Teaching Assistant Experiences

"TA enriched my academic involvement"

“Action speak louder than words” I love helping, so I love teaching I love the feeling of solving problems, so I love teaching I love the sense of satisfaction, so I love teaching I worked ...

Posted by dliu on May 29, 2020

DL Principles

"Tower relies on its foundation"

“Difficult circumstances serve as a textbook of life for people” Deep Learning Principle Standard Scaling Standardization of a dataset is a common requirement for many machine learning es...

Posted by dliu on May 1, 2020

Why LSTM Solves the Gradient Vanishing Problem of RNN

"Genius propose solutions"

“Hope for the best and prepare for the worst” Let’s take a look at why LSTM solves problem of gradient vanishing of RNN. Jump here for details of LSTM avoid gradient vanishing LSTM (Long Sho...

Posted by dliu on April 15, 2020

What is Recurrent Neuron Nets

"Remember what we had!"

“To strive, to seek, to find, and not to yield” RNN is mainly used in the scenario of there is a need to remember the past information and pass propagate to the future. Variants of RNN elimina...

Posted by dliu on April 1, 2020

CNN Architecture Explain and Why it Works

"The utimate goal is to extract dominant features, like brain"

“I insist on the runway, never leave” Convolutional Neural Networks CNN (Convolutional Neural Networks) Why need convolutional from a big picture fully connected deep network requir...

Posted by dliu on March 15, 2020

ML Models Pros vs Cons

"All models are wrong"

“Do one thing at a time, and do well” ML Models & Pros and Cons Error = bias + variance; Tradeoff Supervised vs. unsupervised: Supervised (regression || classification): train model w...

Posted by dliu on March 1, 2020

ML Basic Concepts Review

"Rome was not built in a day"

“The value of life lies not length of days, but in the use of we make of them” This post aims to review some concepts of machine learning, statistics, probability for data scientist interview....

Posted by dliu on February 15, 2020

Donghan (Adrian)'s Blog