The Evaluation Management section within the Portfolio Builder settings allows users to define how the genetic algorithm should manage the evaluation and optimization process over time. These settings help in ensuring that the optimization process remains effective and continues to explore new potential solutions.
Â
Evaluation Management
Â
Start Again When Finished
Description: This setting allows the genetic algorithm to restart automatically after it reaches termination. The process continues in a loop, which can be useful for ongoing optimization, especially in dynamic environments where conditions or requirements may change over time.
- Purpose: Continuous repeating evolution ensures that the optimization process does not stop, potentially finding even better solutions as it keeps running. This is particularly useful in scenarios where the optimal solution can change due to varying market conditions or new data becoming available.
- Use Case: Enable this setting when you want the optimization process to run indefinitely, allowing the system to adapt and find new optimal portfolios continuously.
Â
Restart Evaluation if Fitness Stagnates
Description: This setting restarts the genetic evaluation if the fitness of the solutions does not improve over a specified number of generations. "Fitness stagnation" occurs when the best solutions in the population stop getting better, indicating that the algorithm might be stuck in a local optimum.
- Purpose: Restarting the evaluation when fitness stagnates helps prevent the algorithm from getting trapped in local optima and encourages it to explore new areas of the solution space. This increases the chances of finding a global optimum or better solutions overall.
- Use Case: Set this option to ensure that the algorithm remains effective and continues to improve solutions. Specify the number of generations for which the fitness must stagnate before a restart is triggered. This is useful in maintaining the dynamism and effectiveness of the optimization process.
Â
Summary
The Evaluation Management section in the Portfolio Builder settings provides mechanisms to ensure that the optimization process remains active and effective. By enabling continuous evolution and restarting evaluations upon stagnation, users can ensure that the genetic algorithm consistently seeks better solutions and adapts to changing conditions, enhancing the overall quality and performance of the resulting portfolios.