Abstract: A system may use sensor data from a facility to generate tentative values associated with an event, such as the identification of an item removed from a shelf of the facility. A confidence value associated with each of the tentative values may be less than a confidence threshold. In response, inquiry data seeking confirmation of a tentative value from an associate is generated and sent to one or more associates in the facility. Responses from the associates are collected to determine a selection of one of the tentative values. The selected tentative value is designated as output data for the system. Thereafter, the output data and the original sensor data are designated as training data, which can then be used to train or update machine learning systems. Subsequent use of the updated machine learning systems can yield more accurate results.

Abstract: Described are systems and techniques configured to provide a user interface for an unauthenticated user of a materials handling facility. A transient profile indicative of one or more user preferences may be created and associated with the unauthenticated user. As the unauthenticated user interacts with the facility, preferences may be determined and included in the transient profile. Subsequent operation of the facility may take into account the transient profile. As the unauthenticated user moves throughout the facility, the transient profile may “follow” them, providing an ongoing user experience. Should the unauthenticated user later be authenticated, the transient preferences may be incorporated into a persistent profile.

Abstract: Functionality of sensors on a shelf or other inventory location is tested by encouraging an interaction and determining if the output from the sensors on the shelf is consistent with the interaction. An interaction at the shelf can include, for example, a user adding or removing an item from the shelf. To encourage this type of interaction with the shelf or items therein, an announcement may be generated and presented to a user. The announcement may comprise an advertisement with some type of incentive for the user to add or remove an item from the shelf. Upon verifying such an interaction, sensor data before and after the interaction can be compared to generate diagnostic data for the sensor, which can indicate whether the sensor is operational or malfunctioning.

Abstract: Described is an interactive data processing system configured to facilitate selection by a human associate of tentative values generated by an automated system from sensor data. In one implementation, an event may take place in a materials handling facility. The event may comprise a pick or place of an item from an inventory location, movement of a user, and so forth. The sensor data associated with the event is processed by an automated system to determine tentative values associated with the event. In some situations, an uncertainty may exist as to which of the tentative values accurately reflects the actual event. The sensor data and tentative values may be provided to a human associate. The associate may select one of the tentative values. The selected tentative value and the sensor data may be used to train the automated system.

Abstract: Described are systems and techniques configured to test sensors in a materials handling facility. In one implementation, an age of a user in the facility and a predicted path through the facility may be determined. Physical access controls may be initiated at inventory locations along the predicted path and adjacent to the user. The physical access controls may include deployment of physical barriers, activation of locks, and so forth. As another user who is permitted access approaches a particular inventory location, the physical access controls may be deactivated.