Intuitive entity navigation

A system comprising:

• one or more computers;
• and
• one or more storage devices communicatively coupled to the one or more computers, wherein the one or more storage devices store instructions that, when executed by the one or more computers, cause the one or more computers to perform operations comprising:
• receiving a search query from a user computing device of a user;
• converting the search query into a vector representation of the search query;
• identifying a plurality of n-dimensional entity embeddings in a latent space that are each within a proximity of the vector representation of the search query, wherein each n-dimensional entity embedding represents semantic features of a respective entity and has been generated by operations comprising:
• processing a vector descriptor of the entity using an encoder neural network to generate the n-dimensional entity embedding,
• wherein the encoder neural network has been trained to optimize a cost function using a machine learning training technique;
• projecting, using a projection engine, the respective n-dimensional entity embedding representing each entity into a respective coordinate;
• presenting, on a display of the user computing device, a view of an interactive spatial map of the plurality of entities, wherein each entity is shown at a respective location in the interactive spatial map that is defined by the respective coordinate obtained by projecting the n-dimensional entity embedding of the entity;
• receiving an input from the user computing device that defines an interaction of the user with the interactive spatial map of the plurality of entities;
• responsive to the user interaction, updating the view of the interactive spatial map of the plurality of entities;
• and
• presenting, on the display of the user computing device, the updated view of the interactive spatial map of the plurality of entities.

One or more non-transitory computer storage media storing instructions that when executed by one or more computers cause the one or more computers to perform operations comprising:

• receiving a search query from a user computing device of a user;
• converting the search query into a vector representation of the search query;
• identifying a plurality of n-dimensional entity embeddings in a latent space that are each within a proximity of the vector representation of the search query, wherein each n-dimensional entity embedding represents semantic features of a respective entity and has been generated by operations comprising:
• processing a vector descriptor of the entity using an encoder neural network to generate the n-dimensional entity embedding,
• wherein the encoder neural network has been trained to optimize a cost function using a machine learning training technique;
• projecting, using a projection engine, the respective n-dimensional entity embedding representing each entity into a respective coordinate;
• presenting, on a display of the user computing device, a view of an interactive spatial map of the plurality of entities, wherein each entity is shown at a respective location in the interactive spatial map that is defined by the respective coordinate obtained by projecting the n-dimensional entity embedding of the entity;
• receiving an input from the user computing device that defines an interaction of the user with the interactive spatial map of the plurality of entities;
• responsive to the user interaction, updating the view of the interactive spatial map of the plurality of entities;
• and
• presenting, on the display of the user computing device, the updated view of the interactive spatial map of the plurality of entities.