Multi task learning with incomplete labels for predictive maintenance

A non-transitory computer readable medium, storing instructions for executing a process for multi task learning with incomplete labels for predictive maintenance, the instructions comprising:

• for the data having incomplete labeling to generate a plurality of predictive maintenance models:
• processing the data through a the multi-task learning (MTL) architecture comprising a neural network with generic layers and task specific layers for the plurality of predictive maintenance models configured to conduct tasks to determine outcomes for one or more components associated with the data, each task specific layer corresponding to one of the plurality of predictive maintenance models;
• wherein the generic layers are configured to provide, to the task specific layers, associated data to construct each of the plurality of predictive maintenance models;
• and executing the predictive maintenance models on subsequently recorded data;
• wherein the generic layers comprises long-short term memory (LSTM) layers connected to shared fully connected (FC) layers.

A system for multi task learning with incomplete labels for predictive maintenance, the system comprising:

• one or more components;
• and
• an apparatus configured to receive the data from the one or more components, the apparatus comprising:
• a processor, configured to, for the data having incomplete labeling to generate a plurality of predictive maintenance models:
• process the data through the multi-task learning (MTL) architecture comprising a neural network with generic layers and task specific layers for the plurality of predictive maintenance models configured to conduct tasks to determine outcomes for the one or more components associated with the data, each task specific layer corresponding to one of the plurality of predictive maintenance models;
• wherein the generic layers are configured to provide, to the task specific layers, associated data to construct each of the plurality of predictive maintenance models;
• and execute the predictive maintenance models on subsequently recorded data from the one or more components;
• wherein the generic layers comprises long-short term memory (LSTM) layers connected to shared fully connected (FC) layers.