With AnyLogic partner Skymind, this webinar video demonstrates how the Pathmind platform can ease the introduction of deep reinforcement learning (DRL) capabilities into your AnyLogic projects.
Pathmind enables you to determine optimal policies using the latest DRL code libraries while eliminating the need for complex manual setup and long wait times.
In the webinar recording, you see how Pathmind can:
- import your existing AnyLogic models,
- create and test reward functions,
- automatically apply the latest DRL algorithms – with recommended hyperparameter settings,
- boost compute in the cloud,
- and export policy files ready for you to use back in AnyLogic.
This webinar’s focus is not on teaching deep reinforcement learning, but on how AnyLogic can be used to build simulated training environments and testbeds for use in artificial intelligence.
With AnyLogic AI Integration Lead, Dr. Arash Mahdavi, and Skymind VP of Engineering, Eduardo Gonzalez, the video demonstrates how AnyLogic’s model building capabilities and Pathmind’s automated DRL combine to facilitate simulation-based training.
The webinar is for both AI practitioners who are new to general purpose simulation tools and simulation modelers who are new to DRL.
To find out more about the Pathmind platform or discover more of how AnyLogic works in AI, follow these links:
Webinar video running order:
- Creating observation and action functions that align with desired business outcomes
- Creating and testing DRL reward functions
- Running parallel experiments and evaluating DRL progress
- Using DRL results to validate model assumptions and identify possible corner cases
Many thanks to all who attended this webinar session on automating deep reinforcement learning. Let us know, in the comments below, are you using AI with simulation? What would you like to see in the next webinar?