Adaptive control of exposure. A proportional exposure cap is a maximum fraction applicable to a recipient's total viewable attention in a time window. The total viewable attention represents all viewable advertising content which will be provided to the recipient. A notification of availability of an opportunity to expose a specified recipient to advertising content is received during the time window. The specified recipient's consumed viewable attention is detected. The specified recipient's total viewable attention for the time window is predicted. Responsive to the maximum fraction of the specified recipient's predicted total viewable attention for the time window being greater than the consumed viewable attention of the specified recipient, content is sent to the specified recipient and the consumed viewable attention is updated.

Systems and methods are disclosed for determining an estimate of available user impressions on a network, comprising receiving a request for an estimate of available user impressions for viewing one or more media elements on a network, the request comprising one or more viewer demographic group limitations. A request may be received to include deterministic users and probabilistic users in the estimate of available user impressions. A number of deterministic users may be determined based on query results from a deterministic user data set. A number of probabilistic users may be determined based on query results from a probabilistic user data set, and the estimate of available user impressions may be determined based on the number of deterministic users and the number of probabilistic users.

An automatic system facilitates selection of media properties on which to display an advertisement, responsive to a profile collected on a first media property, where a behavioral-targeting company calculates expected profit for an ad correlated with the profile and arranges for the visitor to be tagged with a tag readable by the selected media property. The profit can be calculated by deducting, from the revenues that are expected to be generated from an ad delivered based on the collected profile, at least the price of ad space at a media property where the BT company might like to deliver ads to the profiled visitor. When the calculated profit is positive (i.e., not a loss), the BT company arranges for the visitor to be tagged with a tag readable by the selected media property through which the BT company expects to profit.

A system to dynamically generate, monitor and optimize an allocation of resources includes a processor-based server to process requests received from the client devices. The processor-based server includes a server processor to generate the allocation of resource sharing one or more display devices selected based on an allocation criterion defined by the information emitter, target rating point goal, a geolocation, a format, and a quality coefficient of each display device in the set of available display devices. The server processor monitors availability of previously unavailable display devices and track real-time changes to the quality coefficients for each display device in the updated set. The server processor dynamically updates and optimizes the allocation of resources based on the updated quality coefficients for the updated set of available display devices.

In the world of digital advertising, optimally allocating an advertisement campaign within a fixed pre-defined budget for an advertising duration aimed at maximizing number of conversions is very important for an advertiser. Embodiments of present disclosure provides a robust and easily generalizable method of optimal allocation of advertisement campaign by formulating it as a constrained Markov Decision Process (MDP) defined by agent state comprising user state and advertiser state, action space comprising a plurality of ad campaigns, state transition routine and a cumulative reward model which rewards maximum total conversions in an advertising duration. The cumulative reward model is trained in conjunction with a deep Q-network for solving the MDP to optimally allocate advertisement campaign for an advertising duration within a constrained budget.

An on-line marketplace provides systems and methods for a mutually beneficial messaging campaign with one or more merchants, in order to promote the merchants' products. One aspect of the messaging campaign includes augmenting the merchants' budgets across multiple commercial channels. The channels may include merchant-operated channels, marketplace-operated channels, and/or third party channels. The marketplace selects when and how to augment the merchants' budgets. The augmented budget comprises an overall content budget for the messaging campaign, including a first budget portion provided by the merchant(s) and a second budget portion provided by the on-line marketplace. The augmented budget may be determined based on a weighted combination of goals of the marketplace and one or more merchants. The augmented budget may be segmented into static and dynamic portions, in which the dynamic portion is varied between channels to determine variables to optimize budget allocation and message performance.

Techniques for managing a multi-marketplace content presentation are described herein. For example, a computer system indicates, via a graphical user interface at a device associated with an account identifier, multiple web sites that offer an item for which content is to be presented. The computer system receives input indicating a selection of a set of web sites of the multiple web sites. The computer system receives a first control and a first target associated with presenting the content in the set of web sites and generates a second control and a second target. The computer system indicates the item and a first metric associated with presenting the content in the web sites based on the first control and the first target. The computer system indicates the selected web site and a second metric associated with presenting the content in the selected web site.

Systems and methods are provided for controlling an online advertising campaign. In one embodiment, a computer-implemented method for controlling an online advertising campaign includes receiving a feedback signal reflecting delivery of the online advertising campaign, and comparing the feedback signal to a delivery reference to generate a campaign level control signal. The method further includes receiving a maximum impression bid price for an inventory unit of the online advertising campaign, the maximum bid price for the at least one inventory unit being set by a user, and calculating, using at least one processor, at least a final bid price based on the maximum bid price, on the campaign level control signal, and on an optimization objective for the online advertising campaign, the optimization objective being set by the user. The method also includes submitting, to an electronic market and based on the calculated final bid price, a bid on an impression from the inventory unit.

Systems and methods are disclosed for optimizing distribution of resources to data elements, comprising receiving a selection of a first objective and a second objective, the first objective and second objective comprising goals associated with distribution of a plurality of data elements; receiving an indication that the first objective has a higher priority than the second objective; receiving a first goal metric associated with the first objective and a second goal metric associated with the second objective; determining a first forecasted metric based on the first goal metric associated with the first objective; determining a second forecasted metric based on the second goal metric associated with the second objective; and allocating resources for the distribution of a plurality of data elements based on the first goal metric, the second goal metric, the first forecasted metric, the second forecasted metric, and the indication that the first objective has a higher priority than the second objective.

A system for training a bidding model comprising: a plurality of tactics stored on at least one database; a plurality of hyperparameters; in response to an available inventory from a publisher relayed through a real time bid server, computing a bid on the available inventory; sending the bid to the real time bid server; receiving an auction result in response to the bid; calculating a plurality of rewards based on the auction result and the tactics; calculate a plurality of q values based on the rewards; calculate a plurality of losses; backpropogating the losses through the bidding model.