Recommending network connections by optimizing for two-sided implicit value of an edge

A method comprising:

• performing, using one or more computer processors of an online network, operations for facilitating establishment of connections in an online network, the operations comprising:
• accessing a set of connection recommendations for a first entity associated with the online network;
• for each connection recommendation in the set of connection recommendations, accessing a ranking value associated with the connection recommendation, determining a utility value corresponding to the connection recommendation, and calculating an adjusted the ranking value for the connection recommendation, the determining of the utility value including applying a machine-learned model to a set of input features identified as being relevant to the utility value, the utility value being a two-sided utility value that includes a prediction of a utility of the first entity combined with a prediction of a utility of a second entity with respect to a key performance indicator of the online network;
• and
• communicating the set of connection recommendations for presentation in an interactive user interface of a client device associated with the first entity in accordance with the adjusted ranking value of each connection recommendation, the interactive user interface including a user interface element that, when activated by the first entity, facilitates establishment of a connection associated with the set of connections.

A non-transitory machine-readable storage medium including instructions that, when executed one or more computer processors, cause the one or more computer processors to perform operations for:

• facilitating establishment of connections in an online network, the operations comprising:
• accessing a set of connection recommendations for a first entity associated with the online network;
• for each connection recommendation in the set of connection recommendations, accessing a ranking value associated with the connection recommendation, determining a utility value corresponding to the connection recommendation, and calculating an adjusted the ranking value for the connection recommendation, the determining of the utility value including applying a machine-learned model to a set of input features identified as being relevant to the utility value, the utility value being a two-sided utility value that includes a prediction of a utility of the first entity combined with a prediction of a utility of a second entity with respect to a key performance indicator of the online network;
• and
• communicating the set of connection recommendations for presentation in an interactive user interface of a client device associated with the first entity in accordance with the adjusted ranking value of each connection recommendation, the interactive user interface including a user interface element that, when activated by the first entity, facilitates establishment of a connection associated with the set of connections.