Implementation Practical Misc
Some implementation magic.
PyTorch Notes
Go Deeper in Convolutions: a Peek
The main aim of conv op is to extract useful features for downstream tasks. And different filters could intuitionally extract different aspect of features via backprop during training. Afterward, all the extracted features are combined to make decisions.
Deciphering AlphaStar on StarCraft II
On 19 December 2018, AlphaStar has decisively beaten human player MaNa with 5-0 on StarCraft II, one of the most challenging Real-Time-Strategy (RTS) games.
Policy Gradient: A Summary !
The mathematical foundations of policy gradient algorithms.
Sorting Algorithms
A summary of sorting algorithms.
Deep Q-Networks: A Summary
A summary of key advances of Deep Q-Networks.
Optimization Methods in Deep Learning
An Introduction to Activation Functions
Activation functions lead to non-linearity in neural networks. Most common types are Sigmoid, Tanh, Relu, etc.