A Disciplined Approach to Neural Network Hyper-Parameters – Paper Dissected

Training a neural network requires carefully selecting hyper-parameters. The optimal parameters vary from one dataset to another. With so many things to tune, this can easily go out of control. Leslie N. Smith in his paper - A Disciplined Approach to Neural Network Hyper-Parameters: Part 1 - Learning Rate, Batch Size, Momentum, and Weight Decay discusses several efficient…

Understanding the Working of Universal Language Model Fine Tuning (ULMFiT)

(Edit) A big thanks to Jeremy Howard for the shout-out 😊 https://twitter.com/jeremyphoward/status/1008156649788325889 Transfer Learning in natural language processing is an area that had not been explored with great success. But, last month (May 2018), Jeremy Howard and Sebastian Ruder came up with the paper - Universal Language Model Fine-tuning for Text Classification which explores the benefits…