Abstract: Technologies are provided for the generating of optimal policies for bidding in auctions having unknown dynamics. In some embodiments, a computing system can configure many multi-armed bandit (MAB) models defining candidate directed contents for a sequence of pages. A particular MAB model of the many MAB models defines candidate directed contents for a particular page in the sequence of pages, where each arm in the particular MAB model corresponds to a candidate impression on the particular page. The computing system can then determine a solution to an optimization problem with respect to an objective function based on an expected long-term reward for a defined impression on the first page, a defined impression on the second page, and a defined impression on the third page. The solution results in respective directed content for presentation on the first, second, and third pages.

Abstract: A modular interaction device includes a wireless identification reading module to receive a unique identifier from a wireless identification tag attached with an object. The unique identifier may be sent to a server. A voice prompt is received by the modular interaction device and at least a portion of the voice prompt is sent to the server, in accordance with at least one embodiment. An audio response may be received from the server and provided to an audio output device of the modular interaction device. The audio response is generated based at least in part on a character profile associated with the unique identifier, in accordance with at least one embodiment. The audio response may be responsive to the voice prompt.