Articles on machine learning in trading

In previous articles, we discussed the Decision Transformer method and several algorithms derived from it. We experimented with different goal setting methods. During the experiments, we worked with various ways of setting goals. However, the model's study of the earlier passed trajectory always remained outside our attention. In this article. I want to introduce you to a method that fills this gap.

This article introduces the Conformer algorithm originally developed for the purpose of weather forecasting, which in terms of variability and capriciousness can be compared to financial markets. Conformer is a complex method. It combines the advantages of attention models and ordinary differential equations.

With the rapid development of artificial intelligence today, language models (LLMs) are an important part of artificial intelligence, so we should think about how to integrate powerful LLMs into our algorithmic trading. For most people, it is difficult to fine-tune these powerful models according to their needs, deploy them locally, and then apply them to algorithmic trading. This series of articles will take a step-by-step approach to achieve this goal.

In the context of reinforcement learning, model procrastination can be caused by several reasons. The article considers some of the possible causes of model procrastination and methods for overcoming them.

In the ever-changing world of trading, adapting to market shifts is not just a choice—it's a necessity. New patterns and trends emerge everyday, making it harder even the most advanced machine learning models to stay effective in the face of evolving conditions. In this article, we’ll explore how to keep your models relevant and responsive to new market data by automatically retraining.

Discover how to fix a critical flaw in financial machine learning that causes overfit models and poor live performance—label concurrency. When using the triple-barrier method, your training labels overlap in time, violating the core IID assumption of most ML algorithms. This article provides a hands-on solution through sample weighting. You will learn how to quantify temporal overlap between trading signals, calculate sample weights that reflect each observation's unique information, and implement these weights in scikit-learn to build more robust classifiers. Learning these essential techniques will make your trading models more robust, reliable and profitable.

Transformer architecture-based models demonstrate high efficiency, but their use is complicated by high resource costs both at the training stage and during operation. In this article, I propose to get acquainted with algorithms that allow to reduce memory usage of such models.

We invite you to get acquainted with the NAFS (Node-Adaptive Feature Smoothing) method, which is a non-parametric approach to creating node representations that does not require parameter training. NAFS extracts features of each node given its neighbors and then adaptively combines these features to form a final representation.

This article discusses the use of the Go-Explore algorithm over a long training period, since the random action selection strategy may not lead to a profitable pass as training time increases.

The article considers the Atomic Orbital Search (AOS) algorithm, which uses the concepts of the atomic orbital model to simulate the search for solutions. The algorithm is based on probability distributions and the dynamics of interactions in the atom. The article discusses in detail the mathematical aspects of AOS, including updating the positions of candidate solutions and the mechanisms of energy absorption and release. AOS opens new horizons for applying quantum principles to computing problems by offering an innovative approach to optimization.

The AI breakthroughs dominating headlines, from ChatGPT to self-driving cars, aren’t built from isolated models but through cumulative knowledge transferred from various models or common fields. Now, this same "learn once, apply everywhere" approach can be applied to help us transform our AI models in algorithmic trading. In this article, we are going to learn how we can leverage the information gained across various instruments to help in improving predictions on others using transfer learning.

In this article we will build a basic neuron. And although it looks simple, and many may consider this code completely trivial and meaningless, I want you to have fun studying this simple sketch of a neuron. Don't be afraid to modify the code, understanding it fully is the goal.

The article considers the algorithm of the MEC family called the simple mind evolutionary computation algorithm (Simple MEC, SMEC). The algorithm is distinguished by the beauty of its idea and ease of implementation.

In this article, we will consider the algorithm that demonstrates the most controversial results of all those discussed previously - the differential evolution (DE) algorithm.

Here I will consider the fairly new Stochastic Marginal Actor-Critic (SMAC) algorithm, which allows building latent variable policies within the framework of entropy maximization.

In this article, we revisit the classic moving average crossover strategy to assess its current effectiveness. Given the amount of time since its inception, we explore the potential enhancements that AI can bring to this traditional trading strategy. By incorporating AI techniques, we aim to leverage advanced predictive capabilities to potentially optimize trade entry and exit points, adapt to varying market conditions, and enhance overall performance compared to conventional approaches.

We will add Deep Learning to those three examples that were published in previous articles and compare results with previous. The aim is to learn how to add DL to other EA.

Moving Average and Stochastic Oscillator are very common indicators that some traders may not use a lot because of their lagging nature. In a 3-part ‘miniseries' that considers the 3 main forms of machine learning, we look to see if this bias against these indicators is justified, or they might be holding an edge. We do our examination in wizard assembled Expert Advisors.

The Learning Rate, is a step size towards a training target in many machine learning algorithms’ training processes. We examine the impact its many schedules and formats can have on the performance of a Generative Adversarial Network, a type of neural network that we had examined in an earlier article.

Category Theory is a diverse and expanding branch of Mathematics which as of yet is relatively uncovered in the MQL5 community. These series of articles look to introduce and examine some of its concepts with the overall goal of establishing an open library that provides insight while hopefully furthering the use of this remarkable field in Traders' strategy development.

The quality of future state predictions plays an important role in the Goal-Conditioned Predictive Coding method, which we discussed in the previous article. In this article I want to introduce you to an algorithm that can significantly improve the prediction quality in stochastic environments, such as financial markets.

With the rapid development of artificial intelligence today, language models (LLMs) are an important part of artificial intelligence, so we should think about how to integrate powerful LLMs into our algorithmic trading. For most people, it is difficult to fine-tune these powerful models according to their needs, deploy them locally, and then apply them to algorithmic trading. This series of articles will take a step-by-step approach to achieve this goal.

We will analyze alternative data curated by the Chicago Board Of Options Exchange (CBOE) to improve the accuracy of our deep neural networks when forecasting the XAUEUR symbol.

In this article, we will take a quick look at some methods to get a function that can represent our data in the database. I will not go into detail about how to use statistics and probability studies to interpret the results. Let's leave that for those who really want to delve into the mathematical side of the matter. Exploring these questions will be critical to understanding what is involved in studying neural networks. Here we will consider this issue quite calmly.

In this article. we will discuss a method of complex multimodal interaction analysis and feature understanding.

In this article, we will continue exploring the Artificial Bee Hive Algorithm (ABHA) by diving into the code and considering the remaining methods. As you might remember, each bee in the model is represented as an individual agent whose behavior depends on internal and external information, as well as motivational state. We will test the algorithm on various functions and summarize the results by presenting them in the rating table.

In this article we continue with our theme in the last of tackling everyday trading indicators viewed in a ‘new’ light. We are handling horizontal composition of natural transformations for this piece and the best indicator for this, that expands on what we just covered, is the double exponential moving average (DEMA).

In this article, we will get acquainted with an interesting algorithm that is built at the intersection of supervised and reinforcement learning methods.

In this article, I would like to introduce you to an interesting trajectory prediction method developed to solve problems in the field of autonomous vehicle movements. The authors of the method combined the best elements of various architectural solutions.

Join us as we empirically analyzed the MACD indicator, to test if applying AI to a strategy, including the indicator, would yield any improvements in our accuracy on forecasting the EURUSD. We simultaneously assessed if the indicator itself is easier to predict than price, as well as if the indicator's value is predictive of future price levels. We will furnish you with the information you need to decide whether you should consider investing your time into integrating the MACD in your AI trading strategies.

As a result of tests performed in previous articles, we came to the conclusion that the optimality of the trained strategy largely depends on the training set used. In this article, we will get acquainted with a fairly simple yet effective method for selecting trajectories to train models.

The article examines the impact of changing the shape of probability distributions on the performance of optimization algorithms. We will conduct experiments using the Smart Cephalopod (SC) test algorithm to evaluate the efficiency of various probability distributions in the context of optimization problems.

In recent articles, we have seen several options for using the Decision Transformer method. The method allows analyzing not only the current state, but also the trajectory of previous states and actions performed in them. In this article, we will focus on using this method in hierarchical models.

As the model is trained based on the experience reproduction buffer, the current Actor policy moves further and further away from the stored examples, which reduces the efficiency of training the model as a whole. In this article, we will look at the algorithm of improving the efficiency of using samples in reinforcement learning algorithms.

Number Walls are a variant of Linear Shift Back Registers that prescreen sequences for predictability by checking for convergence. We look at how these ideas could be of use in MQL5.