Downloads

A Note on the Validity of Cross-Validation for Evaluating Autoregressive Time Series Prediction

In the last report we made a point on the use of Cross validation with time series. If you want to go in depth with the topic, you can read the following note produced by the University of Melbourne.

Archetypal Analysis as an autoencoder

In this article we show how the structure of an autoencoder network can be used to produce the output of an archetypal analysis reducing the computational effort.

Financial Prediction!

This report is the first of a series on machine learning applied to finance, in which we try to highlight the possible use of ML and DL in finance.

A dive into reinforcement learning

Reinforcement learning is a particular area of machine learning concerned with decision making and how software agents take actions in a particular environment in order to maximize a particular function known as reward function.

PCA improvement and generalization with deep learning

In this article we show how the structure of an autoencoder network can be used to produce the output of an archetypal analysis reducing the computational effort.

PERMUTATION IMPORTANCE

Our A.I.4trading team will explain the importance of the feature analysis for the machine learning and deep learning model through a series of three report. In this first paper it will be analyzed a technique known as “permutation importance” which will be applied to Random forest and Extremely randomized Trees models. The paper will explain the theoretical bases of this technique and its strengths compared to feature importance evaluation through the “Gini index” for tree models. There will be then showed experimental results of this technique.