Principles for Faster Transformer Inference

Recursion, Dynamic Programming, and Linear Programming

MLE VS MAP

Verification of Convex Functions using Hessian Matrix

Some Intuitions to Training Generative Adversarial Networks

Approximate Algorithm for Finding Roots to Real-Valued Function

Conventional Classification Loss Functions Elucidated

Model Optimization with Genetically Evolved Hyperparameters

A Review on the Classic Word2Vec Models

Orchestrated Softmax for Fast Language Model Training

Regularization for Classification Models

Accelerate the Training of Neural Language Models