Neural network system and training method thereof, and computer readable medium

Abstract

Theembodiment of that present application provide a neural network system, Among them, the neural network system, includes a clustering module for clustering a plurality of simulation results to determine a plurality of simulation clustering results, and determining a corresponding plurality of action sets according to the plurality of simulation clustering results, wherein each action set includes at least one action sample, and the simulation result is a simulation result of a target entity after the action sample is executed; The reinforcement learning module is configured to select an action set to be executed by the target entity from the plurality of action sets according to a state set of the target entity and output the selected action set to be executed by the target entity. The scheme provided by the embodiment adopts the reinforcement learning module of the action set training neural network system, and the order of magnitude of the action set is greatly reduced compared with the action sample, so that the training time of the neural network system can be shortened, the speed of finding the optimal solution in the training process can be increased, and the accuracy of the training result can be improved.

Description

technical field [0001] The embodiments of the present application relate to the technical field of artificial intelligence, and in particular to a neural network system, a training method thereof, and a computer-readable medium. Background technique [0002] With the development of artificial intelligence technology, how to better use artificial intelligence to assist people to complete their work, or to make intelligent entities such as robots to interact with the environment more flexibly, is an important development direction of artificial intelligence at present. [0003] At present, in order for the artificial intelligence agent to output accurate decisions, it is necessary to train the neural network based on all possible outputs, which needs to rely on huge data support, relying on massive data for decision-making, making it difficult for the algorithm to converge quickly. For example, when training an artificial agent for a vehicle, the vehicle can be gridded, the ch...

AI Technical Summary

Problems solved by technology

However, this model will involve a large number of mesh changes, tens of thousands of mesh changes may occur, and if the mesh division is more detailed, there may even be millions of mesh changes

When using the existing training method, all grid changes need to be trained, resulting in extremely slow training speed, extremely difficult to find the optimal solution during the training process, and a large number of samples lead to a decrease in the accuracy of the training results, which is not conducive to artificial intelligent development

Images