Copy

Newsletter

Hey there, you just received the monthly depends-on-the-definition newsletter for May 2019.

Post of the month: Introduction to n-gram language models
You might have heard, that neural language models power a lot of the recent advances in natural language processing. Namely large models like Bert and GPT-2.
But there is a fairly old approach to language modelling that is quite successful in a way. I always wanted to play with the, so called n-gram language model. So here’s a post about them.
Read More

Paper pick

In this work, the authors exploit the structure of deep learning to enable new learning-based inference and decision strategies that achieve desirable properties such as robustness and interpretability. They introduce the Deep k-Nearest Neighbors (DkNN). This hybrid classifier combines the k-nearest neighbors algorithm with representations of the data learned by each layer of a DNN.

Tips & Tricks

Alibi is an open source Python library aimed at machine learning model inspection and interpretation.

Recommended reading

Initialization can have a significant impact on convergence in training deep neural networks. Simple initialization schemes can accelerate training, but they require care to avoid common pitfalls. In this post by deeplearning.ai, you will learn how to initialize neural network parameters effectively and play with different initialization strategies.
Copyright © 2019 depends-on-the-definition, All rights reserved.