Adversarial Cooperative Imitation Learning for Dynamic Treatment

Abstract

Methods and systems for responding to changing conditions include training a model, using a processor, using trajectories that resulted in a positive outcome and trajectories that resulted in a negative outcome. Training is performed using an adversarial discriminator to train the model to generate trajectories that are similar to historical trajectories that resulted in a positive outcome, and using a cooperative discriminator to train the model to generate trajectories that are dissimilar to historical trajectories that resulted in a negative outcome. A dynamic response regime is generated using the trained model and environment information. A response to changing environment conditions is performed in accordance with the dynamic response regime.

What is claimed is: 
     
         1 . A method for responding to changing conditions, comprising:
 training a model on historical treatment trajectories, using a processor, including trajectories that resulted in a positive health outcome and trajectories that resulted in a negative health outcome, by using an adversarial discriminator to train the model to generate trajectories that are similar to historical treatment trajectories that resulted in a positive health outcome, and by using a cooperative discriminator to train the model to generate trajectories that are dissimilar to historical treatment trajectories that resulted in a negative health outcome, and including iteratively training the adversarial discriminator, the cooperative discriminator, and a dynamic treatment regime using a three-party optimization;   generating the dynamic treatment regime for a patient using the trained model and patient information; and   treating the patient in accordance with the dynamic treatment regime, in a manner that is responsive to changing patient conditions, by triggering one or more medical devices to administer a treatment to the patient.   
     
     
         2 . The method of  claim 1 , wherein the adversarial discriminator, the cooperative discriminator, and the dynamic treatment regime are implemented as multiple-layer perceptrons. 
     
     
         3 . The method of  claim 1 , wherein training the model comprises training an environment model that encodes the patient information as a vector in a latent space. 
     
     
         4 . The method of  claim 3 , wherein the environment model is implemented as a variational auto-encoder network. 
     
     
         5 . The method of  claim 1 , wherein responding to changing patient conditions comprises automatically performing a responsive action to correct a negative condition. 
     
     
         6 . A system for responding to changing conditions, comprising:
 a machine learning model, configured to generate a dynamic treatment regime for using patient information;   a model trainer, configured to train the machine learning model, including trajectories that resulted in a positive health outcome and trajectories that resulted in a negative health outcome, by using an adversarial discriminator to train the machine learning model to generate trajectories that are similar to historical treatment trajectories that resulted in a positive health outcome, and by using a cooperative discriminator to train the model to generate trajectories that are dissimilar to historical treatment trajectories that resulted in a negative health outcome, and to iteratively train the adversarial discriminator, the cooperative discriminator, and a dynamic treatment regime using a three-party optimization; and   a response interface, configured to trigger one or more medical devices to treat the patient in accordance with the dynamic treatment regime in a manner that is responsive to changing patient conditions.   
     
     
         7 . The system of  claim 6 , wherein the adversarial discriminator, the cooperative discriminator, and the dynamic treatment regime are implemented as multiple-layer perceptrons in the machine learning model. 
     
     
         8 . The system of  claim 6 , wherein the model trainer is further configured to train an environment model that encodes the patient information as a vector in a latent space. 
     
     
         9 . The system of  claim 8 , wherein the environment model is implemented as a variational auto-encoder network in the machine learning model. 
     
     
         10 . The system of  claim 6 , wherein the response interface is further configured to automatically perform a responsive action to correct a negative condition.