Methods for building a deep latent feature extractor for industrial sensor data

A system for executing an industrial neural network, wherein the system is configured to:

• execute a latent feature extractor;
• execute a combiner module downstream the latent feature extractor in case multiple input signals are provided to the industrial neural network;
• and
• execute a specific task module that is a sub neural network for the industrial neural network downstream the latent feature extractor and the combiner module, respectively,
• wherein the latent feature extractor is pretrained by:
• providing non-uniform training data comprising training examples for each single task of a multitude of tasks covering different problems in an industrial field;
• and
• optimizing parameters of the industrial neural network of the latent feature extractor based on the multitude of tasks, wherein the following acts are executed iteratively until the parameters have converged:
• randomly selecting a single task from the multitude of tasks for optimizing the parameters;
• sampling the randomly selected task for iterations with the respective training examples of the non-uniform training data for optimizing the parameters based on the randomly selected task,
• wherein each iteration of the iterations comprises:
• extracting latent features from a training signal of the current training example of the respective training examples with the latent feature extractor;
• and
• deriving an output signal for the randomly selected task from the extracted latent features with a respective task module of the randomly selected task.