Methods for controlling use of computing resources, such as virtual game consoles

A method for training an artificial intelligent agent to play a video game on a cloud-based game console shared with human users, comprising:

• providing the artificial intelligent agent to interact in the video game;
• configuring trainers to review experiences from the artificial intelligent agents and improve policies for the artificial intelligent agents for interacting with the video game;
• monitoring a state of the experiment with a resource control service and determining whether to run the experiment once the experiment is in a scheduling state;
• starting the experiment on a predetermined number of the cloud-based game consoles, with a predetermined number of the data gatherers, at one or more predetermined environments;
• providing a resource management service for monitoring an amount of the cloud-based game consoles available at each of the environments, the resource management service including:
• a first resource management module measuring load due to activity of the human users on the cloud-based game consoles at each of the environments;
• a second resource management module predicting future load due to activity of the human users;
• and
• a third resource management module determining the cloud-based game consoles given to the research users;
• reading, by the resource control system. a number of the cloud-based game consoles available;
• identifying, by the resource control system, one or more experiments to suspend;
• and
• moving the one or more experiments identified to suspend into a <suspending> state.

A method for managing computing resources of cloud-based game consoles between human users and research users to ensure a sufficient amount of cloud-based game consoles for the human users, the method comprising:

• providing a resource management service for the computing resources, the resource management service including:
• a first resource management module measuring load due to human activity;
• a second resource management module predicting future load due to human activity;
• and
• a third resource management module determining the computing resources given to the research users;
• and
• providing a resource control service for the computing resources, the resource control service including:
• a first resource control module reading a number of computing resources available;
• and
• a second resource control module identifying one or more experiments to suspend;
• moving the one or more experiments identified to suspend into a <suspending> state;
• assigning the resource control service at each environment containing the cloud-based game consoles, wherein each resource control service monitors for the one or more experiments in the <suspending> state;
• terminating each process under control of the resource control service for the one or more experiments in the <suspending> state;
• saving state information of a trainer running the terminated processes to a remote storage;
• changing a status of each process under control of the resource control service to a <suspended> state when each process under control of the resource control service is terminated;
• and
• changing the one or more experiments to a <suspended> state when each of the resource control service at each environment containing the computing resources has its respective processes for the one or more experiments in the <suspended> state.