Networked system and information processing method

Abstract

The networked system comprises multiple machine learning processing engines. These engines are configured to operate on one or more computing devices. Each engine includes a determining unit for cryptographically identifying other machine learning processing engines. Based on the determined relationships between itself and other machine learning processing engines, each engine exchanges information with the other engines and performs model operations.

Claims

1. A networked system comprising multiple machine learning processing engines, wherein each of the multiple machine learning processing engines is configured to operate on one or more computing devices, each of the multiple machine learning processing engines comprising a determining unit, the determining unit being used to determine other machine learning processing engines by cryptographic methods, and wherein each of the multiple machine learning processing engines exchanges information with the other machine learning processing engines and performs model operations based on the relationship between itself and the determined other machine learning processing engines.

2. The networked system as described in claim 1, wherein, Each of the plurality of machine learning processing engines further comprises a decision unit, which is used to determine the permissions granted to the other machine learning processing engines according to a predetermined setting.

3. The networked system as described in claim 2, wherein, The predetermined settings establish a correspondence between the network addresses of other machine learning processing engines and one or more permissions granted to those other machine learning processing engines.

4. The networked system as described in any one of claims 1 to 3, wherein, The plurality of machine learning processing engines each have settings to establish a correspondence between information representing other machine learning processing engines and network addresses of those other machine learning processing engines.

5. The networked system as described in any one of claims 1 to 3, wherein, The determining unit authenticates the network addresses of the other machine learning processing engines.

6. The networked system as described in any one of claims 1 to 3, wherein, Each of the plurality of machine learning processing engines has a private key for determination by the cryptographic method.

7. The networked system as described in any one of claims 1 to 3, wherein, The model operation includes at least one of the following: data collection, preprocessing of collected data, model training, model evaluation, model deployment, inference using the model, monitoring of model performance, model maintenance, and model retraining.

8. An information processing method, which is an information processing method in a system having multiple machine learning processing engines, wherein the multiple machine learning processing engines are respectively configured to operate on one or more computing devices, the information processing method comprising: a step for each of the multiple machine learning processing engines to determine other machine learning processing engines by cryptographic methods; and a step for each of the multiple machine learning processing engines to exchange information with other machine learning processing engines and perform model operations based on the determined relationship between the other machine learning processing engine and itself.

Images