Abstract: In some examples, a processor may access past order information for a plurality of merchants to determine, for an upcoming time period, a number of orders predicted to be received by the merchants. The processor may determine geographic locations of delivery agent devices, each associated with a respective delivery agent. The processor may compare, for the upcoming time period, a number of active delivery agents with the number of orders predicted to be received. Based at least on determining a predicted shortage of active delivery agents for the upcoming time period, the processor selects a first delivery agent of the plurality of delivery agents to pick up a first order from a first pickup location of a first merchant and a second order from a second pickup location of a second merchant, and sends a message to a first delivery agent device associated with the first delivery agent.

Abstract: Provided are methods for classifying cells, such as T cells, using machine learning methods. The methods can be used to classify different subsets or types of cells in a mixed population of cells.

Abstract: Exemplary embodiments relate to methods, mediums, and systems for moving language models from a server to the client device. Such embodiments may be deployed in an environment where the server is not able to provide modeling services to the clients, such as an end-to-end encrypted (E2EE) environment. Several different techniques are described to address issues of size and complexity reduction, model architecture optimization, model training, battery power reduction, and latency reduction.

Abstract: Exemplary embodiments relate to methods, mediums, and systems for moving language models from a server to the client device. Such embodiments may be deployed in an environment where the server is not able to provide modeling services to the clients, such as an end-to-end encrypted (E2EE) environment. Several different techniques are described to address issues of size and complexity reduction, model architecture optimization, model training, battery power reduction, and latency reduction.

Abstract: Exemplary embodiments relate to methods, mediums, and systems for moving language models from a server to the client device. Such embodiments may be deployed in an environment where the server is not able to provide modeling services to the clients, such as an end-to-end encrypted (E2EE) environment. Several different techniques are described to address issues of size and complexity reduction, model architecture optimization, model training, battery power reduction, and latency reduction.

Abstract: A system and method to insure the security of a remote credit card transaction is disclosed. A merchant webserver generates a transaction interface on a device accessible by a consumer. The transaction interface allows entry of a mobile phone number and a credit card number. A transaction security server receives the mobile phone number and the credit card number. The transaction security server generates a code and sends the generated code to a mobile device corresponding to the entered mobile phone number. The transaction interface generates a code entry screen. The transaction security server verifies the transaction based on the received credit card number if an entered code from the code entry screen matches the generated code, at which point, the transaction is finalized.

Abstract: An assimilator ensures that a multitude of client computers have the correct data stored in their memory. A daemon process continuously produces checksums of the client's files. A master assembles a set of checksums corresponding to files the client should have. Comparing the checksums reveals corrupt files and nonexistent files on the client. The client can download correct files from image servers that store images of partial and complete file systems.

Abstract: An apparatus for weighing and discharging material such as down, feathers, and the like is disclosed. Feathers, down and the like are transported from a storage container to a weighing drum by means of a flow of air which passes horizontally into and through the weighing drum, whereby the feathers are deposited in the drum. When a desired weight has accumulated, the flow of air to the drum is diverted and the accumulation of feathers may be discharged from the drum. Upon completion of the discharge, the flow of air carrying feathers is again directed into the drum.