Dmitriy Gizlyk / Profil

В предыдущих работах мы всегда оценивали текущее состояния окружающей среды. При этом динамика изменения показателей, как таковая, всегда оставалась "за кадром". В данной статье я хочу познакомить Вас с алгоритмом, который позволяет оценить непосредственное изменение данных между 2 последовательными состояниями окружающей среды.

В данной статье я предлагаю Вам познакомиться с алгоритмом GTGAN, который был представлен в январе 2024 года для решения сложных задач по созданию архитектурного макета с ограничениями на граф.

In the previous article, we got acquainted with one of the methods for detecting objects in an image. However, processing a static image is somewhat different from working with dynamic time series, such as the dynamics of the prices we analyze. In this article, we will consider the method of detecting objects in video, which is somewhat closer to the problem we are solving.

In this article, I propose to look at the issue of building a trading strategy from a different angle. We will not predict future price movements, but will try to build a trading system based on the analysis of historical data.

In our models, we often use various attention algorithms. And, probably, most often we use Transformers. Their main disadvantage is the resource requirement. In this article, we will consider a new algorithm that can help reduce computing costs without losing quality.

This article continues the topic of predicting the upcoming price movement. I invite you to get acquainted with the Multi-future Transformer architecture. Its main idea is to decompose the multimodal distribution of the future into several unimodal distributions, which allows you to effectively simulate various models of interaction between agents on the scene.

The models we create are becoming larger and more complex. This increases the costs of not only their training as well as operation. However, the time required to make a decision is often critical. In this regard, let us consider methods for optimizing model performance without loss of quality.

This article introduces a fairly effective method of multi-agent trajectory forecasting, which is able to adapt to various environmental conditions.

We continue to discuss algorithms for training trajectory prediction models. In this article, we will get acquainted with a method called "AutoBots".

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.

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.

In this article, we will get acquainted with an algorithm that uses closed-form policy improvement operators to optimize Agent actions in offline mode.

In offline learning, we use a fixed dataset, which limits the coverage of environmental diversity. During the learning process, our Agent can generate actions beyond this dataset. If there is no feedback from the environment, how can we be sure that the assessments of such actions are correct? Maintaining the Agent's policy within the training dataset becomes an important aspect to ensure the reliability of training. This is what we will talk about in this article.

Since the first articles devoted to reinforcement learning, we have in one way or another touched upon 2 problems: exploring the environment and determining the reward function. Recent articles have been devoted to the problem of exploration in offline learning. In this article, I would like to introduce you to an algorithm whose authors completely eliminated the reward function.

In this article, we continue discussing methods for collecting data into a training set. Obviously, the learning process requires constant interaction with the environment. However, situations can be different.

Models are trained offline using data from a prepared training dataset. While providing certain advantages, its negative side is that information about the environment is greatly compressed to the size of the training dataset. Which, in turn, limits the possibilities of exploration. In this article, we will consider a method that enables the filling of a training dataset with the most diverse data possible.

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

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.

We continue to discuss the family of Decision Transformer methods. From previous article, we have already noticed that training the transformer underlying the architecture of these methods is a rather complex task and requires a large labeled dataset for training. In this article we will look at an algorithm for using unlabeled trajectories for preliminary model training.

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.