Maplebear Inc. (dba Instacart)

An online system may receive, from a content provider, a content presentation campaign that includes one or more objectives. The online system may define a set of one or more policy functions that automatically controls the content presentation campaign. A policy function may control one or more criteria in bidding content slots. The online system may monitor a realized outcome of the content presentation campaign. The online system may apply a reinforcement learning algorithm in adjusting the set of policy functions. The reinforcement learning algorithm adjusts one or more parameters in the set of policy functions to reduce a difference between the realized outcome and the desired outcome set by the content provider. The online system generates an adjusted set of policy functions and uses the adjusted set of policy functions in bidding content slots to present one or more content items provided by the content provider.

An online system generates a template shopping list for a user by accessing a machine learning model trained based on historical order information associated with the user, applying the model to predict likelihoods of conversion for item categories by the user, and populating the template shopping list with one or more item categories based on the predicted likelihoods. The system ranks one or more item types associated with each item category in the template shopping list and determines a set of collection rules associated with one or more item categories/types based on the historical order information. The system generates a suggested shopping list by populating each item category in the template shopping list with one or more item types and a quantity of each item type based on the ranking and rules and sends the suggested shopping list and rules for display to a client device associated with the user.

An automated checkout system modifies received images of machine-readable labels to improve the performance of a label detection model that the system uses to decode item identifiers encoded in the machine-readable labels. For example, the automated checkout system may transform subregions of an image of a machine-readable label to adjust for distortions in the image's depiction of the machine-readable label. Similarly, the automated checkout system may identify readable regions within received images of machine-readable labels and apply a label detection model to those readable regions. By modifying received images of machine-readable labels, these techniques improve on existing computer-vision technologies by allowing for the effective decoding of machine-readable labels based on real-world images using relatively clean training data.

A system or a method for intelligently routing user inquiries to knowledgeable retail shoppers using machine learning. Upon receiving an inquiry from a client device that includes both text and image content, the system applies machine learning models to identify item categories referenced in the text and shown in the image. The system uses an item availability model—trained on historical retailer inventory data—to identify a retailer likely to carry items in the identified categories and transmits suggestion information to the user's device, prompting a user interface that recommends the retailer. The system selects a shopper associated with the retailer who has subject matter expertise in the relevant item categories. Expertise is determined using a machine-learned model trained on labeled data from historical shopper orders. The system then forwards the user inquiry to the expert shopper's device, enabling direct communication and facilitating more accurate, personalized retail assistance.

An online concierge system detects acquired items included among an inventory of a customer and identifies one or more candidate available items from the acquired items based on a predicted perishability of each item and a predicted amount of each item that was used. The system retrieves recipes, matches the item(s) likely to be available to a set of recipes based on their ingredients, and identifies any remaining items for each matched recipe not likely to be available. The system retrieves a set of attributes associated with the customer and the set of recipes and computes a suggestion score for each recipe based on the attributes. The system ranks the recipes based on their scores, identifies one or more recipes for suggesting to the customer based on the ranking, and sends the recipe(s) and any remaining items for each recipe to a client device associated with the customer.

An online concierge system includes sponsored content items in an interface including different slots for displaying content items. A sponsored content item may be displayed in a single slot or in multiple adjacent slots. The online concierge system determines a content score for various sponsored content items indicating a likelihood of a user interacting with a sponsored content item and a position bias for slots in the interface indicating a likelihood of the user interacting with a slot independent of content in the slot. Position biases are different dependent on a number of slots in which a content item is displayed. The online concierge system generates a graph identifying potential placements of sponsored content items in slots by selecting content items in an order according to their content scores. Sponsored content items are positioned in slots according to a path through the graph that has the highest overall expected value.

An online concierge system may determine recommended search terms for a user. The online concierge system may receive a request from a user to view a user interface configured to receive a search query. The online concierge system retrieves long-term activity data including previous search terms entered by the user while searching for items to add to an online shopping cart. For each previous search term, the online concierge system retrieves categorical search terms corresponding to one or more categories to which the previous search term was mapped. The online concierge system determines a set of nearby categorical search terms and sends, for display via a client device, the set of nearby categorical search terms as recommended search terms.

An online concierge system allows a customer to search items offered by a retailer by providing a set of items to the customer based on a search query. To account for varying availability of items at the retailer, the online concierge system modifies rankings in the set of items having less than a threshold predicted availability at the retailer. This reduces a likelihood selection of an item likely to be unavailable at the retailer. To maintain customer confidence in the items selected based on the search results by maintaining visibility of items relevant to the search query, the online concierge system determines how much an item is modified within the set based on search query attributes, item attributes, or customer characteristics. This allows different items to be adjusted different amounts in a set based on the item, as well as the search query for which the item was selected.

An online concierge system provides a client application executed on a client device for customers to generate orders for fulfillment by the online concierge system. If the client device is unable to establish a data connection to a network, the client application locally caches data on the client device for one or more retailers that includes items that have been previously purchased by the customer or that are popular among customers. The customer generates an order through the client application for a retailer based on the locally cached items for the retailer. The online concierge system application generates an encrypted text message based on the order that is transmitted to the online concierge system via short message service (SMS). The online concierge system may also return messages via SMS, which may be presented by the client application.

An online system accesses a two-tower model trained to identify candidate items for presentation to users, in which the model includes an item tower trained to compute item embeddings and a user tower trained to compute user embeddings. The user tower includes a long-term sub-tower trained to compute long-term embeddings for users and a short-term sub-tower trained to compute short-term embeddings for users. The model is trained based on item data associated with items, user data associated with users, and session data associated with user sessions. The system uses the item tower to compute an item embedding for each of multiple candidate items. The system also uses the long-term sub-tower to compute a long-term embedding for a user. The system then receives session data associated with a current session of the user and uses the short-term sub-tower to compute a short-term embedding for the user based on this session data.