Dmitriy Gizlyk / Profile

The article considers methods of encoding initial data in hyperbolic latent space through anisotropic diffusion processes. This helps to more accurately preserve the topological characteristics of the current market situation and improves the quality of its analysis.

In this article, we discuss Directional Diffusion Models that exploit data-dependent anisotropic and directed noise in a forward diffusion process to capture meaningful graph representations.

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.

In the previous last article within this series, we looked at the Atom-Motif Contrastive Transformer (AMCT) framework, which uses contrastive learning to discover key patterns at all levels, from basic elements to complex structures. In this article, we continue implementing AMCT approaches using MQL5.

The Contrastive Transformer is designed to analyze markets both at the level of individual candlesticks and based on entire patterns. This helps improve the quality of market trend modeling. Moreover, the use of contrastive learning to align representations of candlesticks and patterns fosters self-regulation and improves the accuracy of forecasts.

When we use models to analyze the market situation, we mainly focus on the candlestick. However, it has long been known that candlestick patterns can help in predicting future price movements. In this article, we will get acquainted with a method that allows us to integrate both of these approaches.

Self-supervised learning can be an effective way to analyze large amounts of unlabeled data. The efficiency is provided by the adaptation of models to the specific features of financial markets, which helps improve the effectiveness of traditional methods. This article introduces an alternative attention mechanism that takes into account the relative dependencies and relationships between inputs.

We continue the work started in the previous article on building the RefMask3D framework using MQL5. This framework is designed to comprehensively study multimodal interaction and feature analysis in a point cloud, followed by target object identification based on a description provided in natural language.

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

While analyzing the market situation, we divide it into separate segments, identifying key trends. However, traditional analysis methods often focus on one aspect and thus limit the proper perception. In this article, we will learn about a method that enables the selection of multiple objects to ensure a more comprehensive and multi-layered understanding of the situation.

In this article, we will discuss the Mask-Attention-Free Transformer (MAFT) method and its application in the field of trading. Unlike traditional Transformers that require data masking when processing sequences, MAFT optimizes the attention process by eliminating the need for masking, significantly improving computational efficiency.

In this article, we introduce a method for segmenting 3D objects based on Superpoint Transformer (SPFormer), which eliminates the need for intermediate data aggregation. This speeds up the segmentation process and improves the performance of the model.

Effective identification and preservation of the local structure of market data in noisy conditions is a critical task in trading. The use of the Self-Attention mechanism has shown promising results in processing such data; however, the classical approach does not account for the local characteristics of the underlying structure. In this article, I introduce an algorithm capable of incorporating these structural dependencies.

We invite you to get acquainted with a new approach to detecting objects using hypernetworks. A hypernetwork generates weights for the main model, which allows taking into account the specifics of the current market situation. This approach allows us to improve forecasting accuracy by adapting the model to different trading conditions.

In this article, we will talk about algorithms for using attention methods in solving problems of detecting objects in a point cloud. Object detection in point clouds is important for many real-world applications.

We continue to study algorithms for extracting features from a point cloud. In this article, we will get acquainted with the mechanisms for increasing the efficiency of the PointNet method.

Direct point cloud analysis avoids unnecessary data growth and improves the performance of models in classification and segmentation tasks. Such approaches demonstrate high performance and robustness to perturbations in the original data.

We continue studying the Hierarchical Vector Transformer method. In this article, we will complete the construction of the model. We will also train and test it on real historical data.

We invite you to get acquainted with the Hierarchical Vector Transformer (HiVT) method, which was developed for fast and accurate forecasting of multimodal time series.

Understanding agent behavior is important in many different areas, but most methods focus on just one of the tasks (understanding, noise removal, or prediction), which reduces their effectiveness in real-world scenarios. In this article, we will get acquainted with a model that can adapt to solving various problems.