Acron NN

Introduction

Expert adviser Acron NN - A singel layer (perceptron) Neural Networks Forex robot with virtual (hidden) order management. In machine learning, the perceptron is an algorithm for supervised learning of binary classifiers. A binary classifier is a function which can decide whether or not an input, represented by a vector of numbers, belongs to some specific class. It is a type of linear classifier, i.e. a classification algorithm that makes its predictions based on a linear predictor function combining a set of weights with the feature vector.

Acron NN's default settings have been trained under 3 months from 2019/05/01 to 2019/09/13 on EurUsd time frame H1. It's advisable to use default settings only on this pair and time frame! As described, there are four set of weights in Acron NN that needs attention (supervised learning). These are the main inputs to focus on while optimizing Acron NN. More over there's also a virtual order management system which monitor the executed order itself with hidden take profit, stop loss and trailing stop. These virtual inputs should be optimized after the four weighted inputs or leave it completely off and let Acron NN decide when to close the order with use of reversed signal detection which in this case function as a take or stop to the order.

Keep learning period quite short(2-4 months). Optimization preferable on each tick modelling quality but in a recent test i discovered that is also possible and much faster to do it on control points. Re-optimization of Acron NN when the results begin to get very unprofitable. The presence of losses means market has changed, and re-optimization is necessary. Presence of profits means there is no need of any re-optimization as Acron NN still recognizes market patterns. This is not a self learning neural robot and requires your attention!

Supported currency pairs: 6 (including default)

Always use latest .set files! You find them here including optimization guide 'How to learn' Acron NN

nclick="viewGallery_GalleryManager.GoToPrev();"> nclick="viewGallery_GalleryManager.GoToPrev();">

ta itemprop="image" content="https://c.mql5.com/31/372/acron-nn-screen-9126.png" />

nclick="viewGallery_GalleryManager.GoTonext();"> nclick="viewGallery_GalleryManager.GoTonext();">