Advanced artificial intelligence agent for modeling physical interactions
Described herein are advanced artificial intelligence agents for modeling physical interactions. An apparatus to provide an active artificial intelligence (AI) agent includes at least one database to store physical interaction data and compute cluster coupled to the at least one database. The compute cluster automatically obtains physical interaction data from a data collection module without manual interaction, stores the physical interaction data in the at least one database, and automatically trains diverse sets of machine learning program units to simulate physical interactions with each individual program unit having a different model based on the applied physical interaction data.
1. An apparatus to provide an active artificial intelligence (AI) agent comprising:
at least one database to store physical interaction data; and
a compute cluster coupled to the at least one database, the compute cluster to automatically obtain physical interaction data from a data collection module without manual interaction, to store the physical interaction data in the at least one database, automatically training diverse sets of machine learning program units to simulate physical interactions with each individual program unit having a different model based on applied physical interaction data, and to train a master program unit through jointly approximating and modeling behaviors of a whole set of each individual program unit.
2. The apparatus of claim 1 wherein the compute cluster is to apply input including visual input to the master program unit to generate predicted physical interactions based on the training of the individual program units and the master program unit.
3. The apparatus of claim 2 wherein the predicted physical interactions include predicted physical interactions for robotic applications.
4. The apparatus of claim 1 wherein the program units comprise Bayesian program units.
5. The apparatus of claim 4 , wherein:
each of the Bayesian program units has a different deep neural network (DNN) model based on the applied physical interaction data.
6. A method for providing an active artificial intelligence agent comprising:
automatically obtaining, with a training framework, physical interaction data from a data collection module without manual interaction;
storing the physical interaction data in at least one database;
utilizing the training framework having the physical interaction data to automatically train diverse sets of machine learning program units to simulate physical interactions with each individual program unit having a different model based on applied physical interaction data; and
training a master program unit through jointly approximating and modeling behaviors of a whole set of each individual program unit.
7. The method of claim 6 , further comprising:
applying input into the master program unit to generate predicted physical interactions based on the training of the individual program units and the master program unit.
8. The method of claim 7 , wherein:
the predicted physical interactions include predicted physical interactions for robotic applications.
9. The method of claim 8 wherein the program units comprise Bayesian program units.
10. The method of claim 9 wherein each individual Bayesian program unit has a different deep neural network (DNN) model based on the applied physical interaction data.
11. At least one non-transitory machine-readable medium comprising a plurality of instructions, executed on a computing device, to facilitate the computing device to perform one or more operations comprising:
automatically obtaining, with a training framework, physical interaction data from a data collection module without manual interaction;
storing the physical interaction data in at least one database;
utilizing the training framework having the physical interaction data to automatically train diverse sets of machine learning program units to simulate physical interactions with each individual program unit having a different model based on applied physical interaction data; and
training a master program unit through jointly approximating and modeling behaviors of a whole set of each individual program unit.
12. The non-transitory machine-readable medium of claim 11 , further comprising:
applying input into the master program unit to generate predicted physical interactions based on the training of the individual program units and the master program unit.
13. The non-transitory machine-readable medium of claim 12 , wherein:
the predicted physical interactions include predicted physical interactions for robotic applications.
14. The non-transitory machine-readable medium of claim 13 wherein the program units comprise Bayesian program units.
15. The non-transitory machine-readable medium of claim 14 wherein each individual Bayesian program unit has a different deep neural network (DNN) model based on the applied physical interaction data.
16. A system comprising:
a memory to store instructions and physical interaction data; and
a plurality of cores to execute the instructions to automatically obtain physical interaction data from a data collection module without manual interaction, to store the physical interaction data in the memory, to automatically train diverse sets of machine learning program units to simulate physical interactions with each individual program unit having a different model based on applied physical interaction data, and to train a master program unit through jointly approximating and modeling behaviors of a whole set of each individual program unit.
17. The system of claim 16 wherein the plurality of cores is to apply input including visual input to the master program unit to generate predicted physical interactions based on the training of the individual program units and the master program unit.