Discussing the article: "Neural networks made easy (Part 75): Improving the performance of trajectory prediction models"
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
Check out the new article: Neural networks made easy (Part 75): Improving the performance of trajectory prediction models.
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.
Forecasting the trajectory of the upcoming price movement probably plays one of the key roles in the process of constructing trading plans for the desired planning horizon. The accuracy of such forecasts is critical. In an attempt to improve the quality of trajectory forecasting, we complicate our trajectory forecasting models.
However, this process also has another side of the coin. More complex models require more computing resources. This means that the costs of both training models and their operation increase. The cost of model training needs to be taken into account. However, as for operating costs, they can be even more critical. Especially when it comes to real-time trading using market orders in a highly volatile market. In such cases, we look at methods to improve the performance of our models. Ideally, such optimization should not affect the quality of future trajectory forecasts.
The authors of the method propose to use a simple but powerful map preprocessing algorithm, where the trajectory of the target agent is initially filtered. Then they compute the feasible area where the target agent can interact, taking into account only the geometric information of the map.
Author: Dmitriy Gizlyk