In this article, I introduce an innovative trading algorithm that combines evolutionary algorithms with deep reinforcement learning for Forex trading. The algorithm uses the mechanism of extinction of inefficient individuals to optimize the trading strategy.
We invite you to explore a framework that combines wavelet transforms and a multi-task self-attention model, aimed at improving the responsiveness and accuracy of forecasting in volatile market conditions. The wavelet transform allows asset returns to be decomposed into high and low frequencies, carefully capturing long-term market trends and short-term fluctuations.
We have built a robust feature engineering pipeline using proper tick-based bars to eliminate data leakage and solved the critical problem of labeling with meta-labeled triple-barrier signals. This installment covers the advanced labeling technique, trend-scanning, for adaptive horizons. After covering the theory, an example shows how trend-scanning labels can be used with meta-labeling to improve on the classic moving average crossover strategy.
This piece follows up ‘Part-80’, where we examined the pairing of Ichimoku and the ADX under a Reinforcement Learning framework. We now shift focus to Inference Learning. Ichimoku and ADX are complimentary as already covered, however we are going to revisit the conclusions of the last article related to pipeline use. For our inference learning, we are using the Beta algorithm of a Variational Auto Encoder. We also stick with the implementation of a custom signal class designed for integration with the MQL5 Wizard.
We continue our examination of the StockFormer hybrid trading system, which combines predictive coding and reinforcement learning algorithms for financial time series analysis. The system is based on three Transformer branches with a Diversified Multi-Head Attention (DMH-Attn) mechanism that enables the capturing of complex patterns and interdependencies between assets. Previously, we got acquainted with the theoretical aspects of the framework and implemented the DMH-Attn mechanisms. Today, we will talk about the model architecture and training.
We will look at price discretization methods using Python + MQL5. In this article, I will share my practical experience developing a Python library that implements a wide range of approaches to bar formation — from classic Volume and Range bars to more exotic methods like Renko and Kagi. We will consider three-line breakout candles and range bars analyzing their statistics and trying to define how else the prices can be represented discretely.
This article walks the reader through a reimagined version of the classical Bollinger Band breakout strategy. It identifies key weaknesses in the original approach, such as its well-known susceptibility to false breakouts. The article aims to introduce a possible solution: the Double Bollinger Band trading strategy. This relatively lesser known approach supplements the weaknesses of the classical version and offers a more dynamic perspective on financial markets. It helps us overcome the old limitations defined by the original rules, providing traders with a stronger and more adaptive framework.
The article considers a new population optimization algorithm - Cyclic Parthenogenesis Algorithm (CPA), inspired by the unique reproductive strategy of aphids. The algorithm combines two reproduction mechanisms — parthenogenesis and sexual reproduction — and also utilizes the colonial structure of the population with the possibility of migration between colonies. The key features of the algorithm are adaptive switching between different reproductive strategies and a system of information exchange between colonies through the flight mechanism.
This article follows up ‘Part-74’, where we examined the pairing of Ichimoku and the ADX under a Supervised Learning framework, by moving our focus to Reinforcement Learning. Ichimoku and ADX form a complementary combination of support/resistance mapping and trend strength spotting. In this installment, we indulge in how the Twin Delayed Deep Deterministic Policy Gradient (TD3) algorithm can be used with this indicator set. As with earlier parts of the series, the implementation is carried out in a custom signal class designed for integration with the MQL5 Wizard, which facilitates seamless Expert Advisor assembly.
Machine learning is often viewed through statistical or linear algebraic lenses, but this article emphasizes a geometric perspective of model predictions. It demonstrates that models do not truly approximate the target but rather map it onto a new coordinate system, creating an inherent misalignment that results in irreducible error. The article proposes that multi-step predictions, comparing the model’s forecasts across different horizons, offer a more effective approach than direct comparisons with the target. By applying this method to a trading model, the article demonstrates significant improvements in profitability and accuracy without changing the underlying model.
The article describes the experience of developing a hybrid trading system that combines classical technical analysis with neural networks. The author provides a detailed analysis of the system architecture from basic pattern analysis and neural network structure to the mechanisms behind trading decisions, and shares real code and practical observations.
This article presents a study of the interaction of different activation functions with optimization algorithms in the context of neural network training. Particular attention is paid to the comparison of the classical ADAM and its population version when working with a wide range of activation functions, including the oscillating ACON and Snake functions. Using a minimalistic MLP (1-1-1) architecture and a single training example, the influence of activation functions on the optimization is isolated from other factors. The article proposes an approach to manage network weights through the boundaries of activation functions and a weight reflection mechanism, which allows avoiding problems with saturation and stagnation in training.
In this piece, we look at a key data preparation step for machine learning that is gaining rapid significance. Data Preprocessing Pipelines. These in essence are a streamlined sequence of data transformation steps that prepare raw data before it is fed to a model. As uninteresting as this may initially seem to the uninducted, this ‘data standardization’ not only saves on training time and execution costs, but it goes a long way in ensuring better generalization. In this article we are focusing on some SCIKIT-LEARN preprocessing functions, and while we are not exploiting the MQL5 Wizard, we will return to it in coming articles.
The article describes an innovative approach to forecasting price movements in financial markets using quantum computing. The main focus is on the application of the Quantum Phase Estimation (QPE) algorithm to find prototypes of price patterns allowing traders to significantly speed up the market data analysis.
In this article, we will discuss the hybrid trading system StockFormer, which combines predictive coding and reinforcement learning (RL) algorithms. The framework uses 3 Transformer branches with an integrated Diversified Multi-Head Attention (DMH-Attn) mechanism that improves on the vanilla attention module with a multi-headed Feed-Forward block, allowing it to capture diverse time series patterns across different subspaces.
In the previous article, we introduced the multi-agent adaptive framework MASAAT, which uses an ensemble of agents to perform cross-analysis of multimodal time series at different data scales. Today we will continue implementing the approaches of this framework in MQL5 and bring this work to a logical conclusion.
Dynamic mode decomposition (DMD) is a technique usually applied to high-dimensional datasets. In this article, we demonstrate the application of DMD on univariate time series, showing its ability to characterize a series as well as make forecasts. In doing so, we will investigate MQL5's built-in implementation of dynamic mode decomposition, paying particular attention to the new matrix method, DynamicModeDecomposition().
The article presents the Big Bang - Big Crunch method, which has two key phases: cyclic generation of random points and their compression to the optimal solution. This approach combines exploration and refinement, allowing us to gradually find better solutions and open up new optimization opportunities.
We introduce the Multi-Agent Self-Adaptive Portfolio Optimization Framework (MASAAT), which combines attention mechanisms and time series analysis. MASAAT generates a set of agents that analyze price series and directional changes, enabling the identification of significant fluctuations in asset prices at different levels of detail.
This article takes a fresh perspective on a hidden, geometric source of error that quietly shapes every prediction your models make. By rethinking how we measure and apply machine learning forecasts in trading, we reveal how this overlooked perspective can unlock sharper decisions, stronger returns, and a more intelligent way to work with models we thought we already understood.
The Black Hole Algorithm (BHA) uses the principles of black hole gravity to optimize solutions. In this article, we will look at how BHA attracts the best solutions while avoiding local extremes, and why this algorithm has become a powerful tool for solving complex problems. Learn how simple ideas can lead to impressive results in the world of optimization.
In the previous article, we introduced the multi-agent self-adaptive framework MASA, which combines reinforcement learning approaches and self-adaptive strategies, providing a harmonious balance between profitability and risk in turbulent market conditions. We have built the functionality of individual agents within this framework. In this article, we will continue the work we started, bringing it to its logical conclusion.
We are going to develop a modular trading system that combines Python for data analysis with MQL5 for trade execution. Four independent modules monitor different market aspects in parallel: volumes, arbitrage, economics and risks, and use RandomForest with 400 trees for analysis. Particular emphasis is placed on risk management, since even the most advanced trading algorithms are useless without proper risk management.
Continuing tries to decipher price movements... What about linguistic analysis of the "market dictionary" that we get by converting the binary price code to BIP39? In this article, we will delve into an innovative approach to exchange data analysis and consider how modern natural language processing techniques can be applied to the market language.
I invite you to get acquainted with the Multi-Agent Self-Adaptive (MASA) framework, which combines reinforcement learning and adaptive strategies, providing a harmonious balance between profitability and risk management in turbulent market conditions.
The article provides a detailed discussion of the key components and innovations of the ATA optimization algorithm, which is an evolutionary method with a unique dual behavior system that adapts depending on the situation. ATA combines individual and social learning while using crossover for explorations and migration to find solutions when stuck in local optima.
This article presents an innovative approach to technical analysis based on converting price movements into binary code. The author demonstrates how various aspects of market behavior — from simple price movements to complex patterns — can be encoded in a sequence of zeros and ones.
In this second installment of the MetaTrader 5 Machine Learning Blueprint series, you’ll discover why simple labels can lead your models astray—and how to apply advanced techniques like the Triple-Barrier and Trend-Scanning methods to define robust, risk-aware targets. Packed with practical Python examples that optimize these computationally intensive techniques, this hands-on guide shows you how to transform noisy market data into reliable labels that mirror real-world trading conditions.
In the last piece, we concluded our look at the pairing of the gator oscillator and the accumulation/distribution oscillator when used in their typical setting of the raw signals they generate. These two indicators are complimentary as trend and volume indicators, respectively. We now follow up that piece, by examining the effect that supervised learning can have on enhancing some of the feature patterns we had reviewed. Our supervised learning approach is a CNN that engages with kernel regression and dot product similarity to size its kernels and channels. As always, we do this in a custom signal class file that works with the MQL5 wizard to assemble an Expert Advisor.
This article explores the powerful role of matrix factorization in algorithmic trading, specifically within MQL5 applications. From regression models to multi-target classifiers, we walk through practical examples that demonstrate how easily these techniques can be integrated using built-in MQL5 functions. Whether you're predicting price direction or modeling indicator behavior, this guide lays a strong foundation for building intelligent trading systems using matrix methods.
In the previous work, we discussed the theoretical aspects of the PSformer framework, which includes two major innovations in the classical Transformer architecture: the Parameter Shared (PS) mechanism and attention to spatio-temporal segments (SegAtt). In this article, we continue the work we started on implementing the proposed approaches using MQL5.
Harness the full potential of your MetaTrader 5 terminal by leveraging Python’s data-science ecosystem and the official MetaTrader 5 client library. This article demonstrates how to authenticate and stream live tick and minute-bar data directly into Parquet storage, apply sophisticated feature engineering with Ta and Prophet, and train a time-aware Gradient Boosting model. We then deploy a lightweight Flask service to serve trade signals in real time. Whether you’re building a hybrid quant framework or enhancing your EA with machine learning, you’ll walk away with a robust, end-to-end pipeline for data-driven algorithmic trading.
This article introduces the new PSformer framework, which adapts the architecture of the vanilla Transformer to solving problems related to multivariate time series forecasting. The framework is based on two key innovations: the Parameter Sharing (PS) mechanism and the Segment Attention (SegAtt).
This part focuses on building a flexible, adaptive trading model trained on historical XAUUSD data, preparing it for ONNX export and potential integration into live trading systems.
SAMformer offers a solution to the key drawbacks of Transformer models in long-term time series forecasting, such as training complexity and poor generalization on small datasets. Its shallow architecture and sharpness-aware optimization help avoid suboptimal local minima. In this article, we will continue to implement approaches using MQL5 and evaluate their practical value.
Historical data is far from “trash”—it’s the foundation of any robust market analysis. In this article, we’ll take you step‑by‑step from collecting that history to using it to train a predictive model, and finally deploying that model for live price forecasts. Read on to learn how!
The article presents a simple and accessible way to use a neural network in a trading EA that does not require deep knowledge of machine learning. The method eliminates the target function normalization, as well as overcomes "weight explosion" and "network stall" issues offering intuitive training and visual control of the results.
Discovering a new world of automated trading on 3D bars. What does a trading robot look like on multidimensional price bars? Are "yellow" clusters of 3D bars able to predict trend reversals? What does multidimensional trading look like?
Have you ever missed a sudden market spike or been caught off‑guard when one occurred? The best way to anticipate live events is to learn from historical patterns. Intending to train an ML model, this article begins by showing you how to create a script in MetaTrader 5 that ingests historical data and sends it to Python for storage—laying the foundation for your spike‑detection system. Read on to see each step in action.
