How to compress your machine learning model ? – Knowledge distillation
Let us suppose that you are developing a machine learning model which has to generalize as well as possible on unseen data. In general,...
How to evaluate your translation model ?
In order to evaluate models, we need metrics designed to assess their quality. For most of AI related problems, it is often possible to...
Sequence-to-Sequence architectures
Sequence-to-sequence is a sequence modeling architecture relying on an encoder-decoder mechanism. The encoder receives an input sequence which it encodes into a hidden state...
You work with text? Learn how to preprocess it !
Let us suppose that you receive an NLP task where you have to work with textual data which is highly unstructured and may be...
How to detect Drift in your data ?
IntroductionIn most machine learning problems, we assume that, for each feature, the examples are drawn at random from the same distribution which might not...
Playing the Beer Game using Reinforcement Learning
The Beer Game is a well-known simulation of the bullwhip effect. It consists in building a multi-agent environment which represents the supply chain network:...
The Transformer Network
The Transformer network was presented during the NIPS conference in 2017. It is based on an encoder-decoder paradigm where an input sentence is encoded...
PROPHET : Are you ready to predict stock prices?
Let us agree on one thing: we've all dreamed of predicting the stock prices to invest our money but unfortunately it isn't a simple...
Is the borrower gonna pay?
Lending Club is an American participative finance company based in San Fransisco. It was founded by Renaud Laplanche in 2006. It is a platform...
How does Reinforcement Learning work? – Part I
After talking about supervised learning, where the feedback is immediate (labels provided for every input) in this article, and about unsupervised learning, where there...
