Model aggregation device and model aggregation system
A model aggregation device includes a communication device able to communicate with a plurality of vehicles in which neural network models are learned, a storage device storing a part of the neural network models sent from the plurality of vehicles, and a control device. The neural network model outputs at least one output parameter from a plurality of input parameters. The control device is configured to, if receiving a new neural network model from one vehicle among the plurality of vehicles through the communication device, compare ranges of the plurality of input parameters which were used for learning the new neural network model and ranges of the plurality of input parameters which were used for learning a current neural network model stored in the storage device to thereby determine whether to replace the current neural network model with the new neural network model.
1. A model aggregation device, comprising:
a communication device configured to communicate with a plurality of vehicles in which neural network models are learned by the plurality of vehicles;
a non-transitory memory storing a part of the neural network models sent from the plurality of vehicles; and
a control device, wherein
a neural network model of the neural network models is configured to output at least one output parameter from a plurality of input parameters, and
the control device is configured to, in response to receiving a new neural network model from one vehicle among the plurality of vehicles through the communication device,
compare first ranges of the plurality of input parameters which were used for learning the new neural network model and second ranges of the plurality of input parameters which were used for learning a current neural network model stored in the non-transitory memory to thereby determine whether to replace the current neural network model with the new neural network model, wherein in comparing the first ranges and the second ranges, corresponding types of the plurality of input parameters of the new neural network model and the current neural network model are compared.
2. The model aggregation device according to claim 1 , wherein
the control device is configured to replace the current neural network model with the new neural network model in response to
the first ranges being broader than the second ranges,
a first number of combinations of the plurality of input parameters which were used for learning the new neural network model being smaller than a second number of combinations of the plurality of input parameters which were used for learning the current neural network model, and
first date and time when the new neural network model finishes being learned being later than second date and time when the current neural network model finishes being learned.
3. The model aggregation device according to claim 1 , wherein
the control device is configured to replace the current neural network model with the new neural network model in response to
the first ranges being narrower than the second ranges,
a first number of combinations of the plurality of input parameters which were used for learning the new neural network model being greater than a second number of combinations of the plurality of input parameters which were used for learning the current neural network model, and
first date and time when the new neural network model finishes being learned being later than second date and time when the current neural network model finishes being learned.
4. The model aggregation device according to claim 1 , wherein
the control device is configured to replace the current neural network model with the new neural network model in response to the first ranges being broader than the second ranges.
5. The model aggregation device according to claim 1 , wherein
the control device is configured to replace the current neural network model with the new neural network model in response to
the first ranges being equal to the second ranges, and
a first number of combinations of the plurality of input parameters which were used for learning the new neural network model being greater than a second number of combinations of the plurality of input parameters which were used for learning the current neural network model.
6. The model aggregation device according to claim 1 , wherein
the control device is configured not to replace the current neural network model with the new neural network model in response to a value of at least one of the plurality of input parameters which were used for learning the new neural network model being outside of a predetermined range.
7. A model aggregation system, comprising:
a server; and
a plurality of vehicles, wherein
each of the plurality of vehicles comprises:
a first communication device configured to communicate with the server and
a first control device configured to perform learning of a neural network model outputting at least one output parameter from a plurality of input parameters,
the server comprises:
a second communication device configured to communicate with the plurality of vehicles,
a non-transitory memory storing a part of the neural network model of neural network models sent from the plurality of the vehicles, and
a second control device, and
the second control device is configured to, in response to receiving a new neural network model from one vehicle among the plurality of vehicles through the first communication device and the second communication device, compare first ranges of the plurality of input parameters which were used for learning the new neural network model and second ranges of the plurality of input parameters which were used for learning a current neural network model stored in the non-transitory memory to thereby determine whether to replace the current neural network model with the new neural network model, wherein in comparing the first ranges and the second ranges, corresponding types of the plurality of input parameters of the new neural network model and the current neural network model are compared.