Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for determining, for each of one or more categorical features, a respective vocabulary of categorical feature values of the categorical feature that should be active during processing of inputs by a machine learning model. In one aspect, a method comprises: generating a batch of output sequences, each output sequence in the batch specifying, for each of the categorical features, a respective vocabulary of categorical feature values of the categorical feature that should be active; for each output sequence in the batch, determining a performance metric of the machine learning model on a machine learning task after the machine learning model has been trained to perform the machine learning task with only the respective vocabulary of categorical feature values of each categorical feature specified by the output sequence being active.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for determining, for each of one or more categorical features, a respective vocabulary of categorical feature values of the categorical feature that should be active during processing of inputs by a machine learning model. In one aspect, a method comprises: generating a batch of output sequences, each output sequence in the batch specifying, for each of the categorical features, a respective vocabulary of categorical feature values of the categorical feature that should be active; for each output sequence in the batch, determining a performance metric of the machine learning model on a machine learning task after the machine learning model has been trained to perform the machine learning task with only the respective vocabulary of categorical feature values of each categorical feature specified by the output sequence being active.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for determining, for each of one or more categorical features, a respective vocabulary of categorical feature values of the categorical feature that should be active during processing of inputs by a machine learning model. In one aspect, a method comprises: generating a batch of output sequences, each output sequence in the batch specifying, for each of the categorical features, a respective vocabulary of categorical feature values of the categorical feature that should be active; for each output sequence in the batch, determining a performance metric of the machine learning model on a machine learning task after the machine learning model has been trained to perform the machine learning task with only the respective vocabulary of categorical feature values of each categorical feature specified by the output sequence being active.

Abstract: The present disclosure is directed to systems and methods that include a machine-learned label embedding model that generates feature-based label embeddings for labels in real-time, in furtherance, for example, of selection of labels relative to a particular entity. In particular, one example computing system includes both a machine-learned entity embedding model configured to receive and process entity feature data descriptive of an entity to generate an entity embedding for the entity and a machine-learned label embedding model configured to receive and process first label feature data associated with a first label to generate a first label embedding for the first label.

Abstract: The present disclosure is directed to systems and methods that include a machine-learned label embedding model that generates feature-based label embeddings for labels in real-time, in furtherance, for example, of selection of labels relative to a particular entity. In particular, one example computing system includes both a machine-learned entity embedding model configured to receive and process entity feature data descriptive of an entity to generate an entity embedding for the entity and a machine-learned label embedding model configured to receive and process first label feature data associated with a first label to generate a first label embedding for the first label.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for determining, for each of one or more categorical features, a respective vocabulary of categorical feature values of the categorical feature that should be active during processing of inputs by a machine learning model. In one aspect, a method comprises: generating a batch of output sequences, each output sequence in the batch specifying, for each of the categorical features, a respective vocabulary of categorical feature values of the categorical feature that should be active; for each output sequence in the batch, determining a performance metric of the machine learning model on a machine learning task after the machine learning model has been trained to perform the machine learning task with only the respective vocabulary of categorical feature values of each categorical feature specified by the output sequence being active.

Abstract: A bone plate assembly utilizing at least one bone screw for fixation of adjacent bones of a spine including a base plate having at least one aperture extending therethrough and screw retaining mechanism mounted and movable on the plate between locked and unlocked positions relative to the aperture for preventing the bone screw from backing out from the base plate.

Abstract: A bone plate assembly utilizing at least one bone screw for fixation of adjacent bones of a spine including a base plate having at least one aperture extending therethrough and screw retaining mechanism mounted and movable on the plate between locked and unlocked positions relative to the aperture for preventing the bone screw from backing out from the base plate. Furthermore, a device for placement into an aperture of a base plate including a screw retaining mechanism mountable and movable on a plate between locked and unlocked positions relative to an aperture for preventing a bone screw from backing out from the base plate.

Abstract: The present invention provides for a bone plate assembly utilizing at least one bone screw for fixation of adjacent bones of a spine including a base plate having at least one aperture extending therethrough and screw retaining mechanism mounted and movable on the plate between locked and unlocked positions relative to the aperture for preventing the bone screw from backing out from the base plate. The present invention also provides for a device for placement into an aperture of a base plate including a screw retaining mechanism mountable and movable on a plate between locked and unlocked positions relative to an aperture for preventing a bone screw from backing out from the base plate. Additionally, the present invention provides for a snap ring for placement into an aperture of a base plate including a screw retaining mechanism mountable and movable on the plate between locked and unlocked positions relative to the aperture for preventing a screw from backing out.

Abstract: A bone plate assembly is disclosed utilizing at least one bone screw for fixation of adjacent bones of a spine including a base plate having at least one aperture extending therethrough and screw retaining mechanism mounted and movable on the plate between locked and unlocked positions relative to the aperture for preventing the bone screw from backing out from the base plate. A device is disclosed for placement into an aperture of a base plate including a screw retaining mechanism mountable and movable on a plate between locked and unlocked positions relative to an aperture for preventing a bone screw from backing out from the base plate. A snap ring for placement into an aperture of a base plate including a screw retaining mechanism mountable and movable on the plate between locked and unlocked positions relative to the aperture for preventing a screw from backing out is disclosed.

Abstract: The present invention provides for a bone plate assembly utilizing at least one bone screw for fixation of adjacent bones of a spine including a base plate having at least one aperture extending therethrough and screw retaining mechanism mounted and movable on the plate between locked and unlocked positions relative to the aperture for preventing the bone screw from backing out from the base plate. The present invention also provides for a device for placement into an aperture of a base plate including a screw retaining mechanism mountable and movable on a plate between locked and unlocked positions relative to an aperture for preventing a bone screw from backing out from the base plate. Additionally, the present invention provides for a snap ring for placement into an aperture of a base plate including a screw retaining mechanism mountable and movable on the plate between locked and unlocked positions relative to the aperture for preventing a screw from backing out.

Abstract: The present invention provides for a bone plate assembly utilizing at least one bone screw for fixation of adjacent bones of a spine including a base plate having at least one aperture extending therethrough and screw retaining mechanism mounted and movable on the plate between locked and unlocked positions relative to the aperture for preventing the bone screw from backing out from the base plate. The present invention also provides for a device for placement into an aperture of a base plate including a screw retaining mechanism mountable and movable on a plate between locked and unlocked positions relative to an aperture for preventing a bone screw from backing out from the base plate. Additionally, the present invention provides for a snap ring for placement into an aperture of a base plate including a screw retaining mechanism mountable and movable on the plate between locked and unlocked positions relative to the aperture for preventing a screw from backing out.

Abstract: The present invention pertains to a system for sending ATM cells. The system comprises an ATM network along which ATM cells travel. The system comprises source nodes connected to the ATM network which produce packets having cells and transmit the cells on the ATM network. The system comprises I input switches connected to the ATM network, where I is greater than or equal to 1 and is an integer, which receive cells from the ATM network transmitted by the source node. The system comprises M intermediate switches connected to the ATM network, where M is greater than or equal to 1, which receive signals from the ATM network transmitted by at least one input switch. The intermediate switch discards cells only of a same packet when a predetermined event occurs so the discarded cells are not transmitted by the intermediate switch. The system comprises destination nodes connected to the ATM network which receive non-discarded ATM cells from the intermediate switch.