Machine learning technique for automatic modeling of multiple-valued outputs

A system, comprising:

• a processor configured to execute a model that implements a machine-learning algorithm;
• and
• a memory that stores a database including a plurality of latent descriptor vectors,
• wherein the model is trained by:
• receiving a set of training data, wherein the set of training data includes a plurality of input vectors, a plurality of training identifiers, and a plurality of desired output vectors, wherein each input vector in the plurality of input vectors is associated with a training identifier in the plurality of training identifiers, initializing, prior to training the model, the plurality of latent descriptor vectors, wherein each latent descriptor vector stored in the database is mapped to one training identifier in the plurality of training identifiers, providing a plurality of augmented input vectors as inputs to the model, wherein each augmented input vector in the plurality of augmented input vectors comprises an input vector in the plurality of input vectors and a latent descriptor vector selected from the database using the training identifier that is associated with the input vector,
• applying, by the model, a set of model parameters to the plurality of augmented input vectors to generate a set of output vectors, and
• adjusting the set of model parameters and at least one latent descriptor vector stored in the database based on differences between each output vector in the set of output vectors and a corresponding desired output vector in the plurality of desired output vectors.