Abstract: Methods, systems, and devices for wireless communications are described. In a wireless communications system, a network entity may determine to transmit first downlink control information (DCI) associated with a demodulation reference signal (DMRS) allocated to multiple resource elements of a control channel. The network entity may select a CDM group used for encoding second (e.g., future) DCI, for example, based on identifying a phase change pattern associated with the CDM group. A user equipment (UE) may monitor for the DMRS during a first time interval, and determine the CDM group for encoding the second DCI, for example using blind decoding. In some cases, the UE may receive the second DCI via the control channel based on the CDM group determined in association with the second DCI.

Abstract: Methods, systems, and devices for dynamic physical downlink shared channel (PDSCH) mapping modes are described. In some examples, a user equipment (UE) may receive a first control message identifying one or more sets of resources around which the UE may rate match for a downlink channel. In some examples, the UE may receive a second control message including an indication that the UE may perform either rate matching around at least one set of resources of the one or more sets of resources or receiving the downlink channel on the at least one set of resources. As such, the UE may receive signals on the downlink channel according to the indication received in the second control message.

Abstract: An example system includes a processor to receive data for a multi-objective task. The processor is to also perform the multi-objective task on the received data via a trained primal network. The primal network and a dual network are trained for a multi-objective task using a Lagrangian loss function representing a number of objectives. The primal network is trained to minimize the Lagrangian loss function and the dual network is trained to maximize the Lagrangian loss function.

Abstract: An example system includes a processor to receive a prefix of conversation and a text input. The processor is to also generate a completed response based on the prefix of conversation and the text input via a trained primal network. The primal network is trained to minimize a Lagrangian loss function representing a number of objectives and a dual network is trained to maximize the Lagrangian loss function.

Abstract: A method, computer system, and a computer program product for enhanced rating predictions is provided. The present invention may include receiving a user input. The present invention may then include translating the received user input into an embedding matrix and inputting the embedding matrix into a deep neural network. The present invention may further include generating, by the deep neural network, an output vector, a user bias term and an item bias term based on the embedding matrix. The present invention may then include calculating a predicted rating based on the generated output vector, the generated user bias term and the generated item bias term. The present invention may then include determining an accuracy of the predicted rating.

Abstract: An example system includes a processor to receive data for a multi-objective task. The processor is to also perform the multi-objective task on the received data via a trained primal network. The primal network and a dual network are trained for a multi-objective task using a Lagrangian loss function representing a number of objectives. The primal network is trained to minimize the Lagrangian loss function and the dual network is trained to maximize the Lagrangian loss function.

Abstract: An example system includes a processor to receive a prefix of conversation and a text input. The processor is to also generate a completed response based on the prefix of conversation and the text input via a trained primal network. The primal network is trained to minimize a Lagrangian loss function representing a number of objectives and a dual network is trained to maximize the Lagrangian loss function.

Abstract: A method, computer system, and a computer program product for enhanced rating predictions is provided. The present invention may include receiving a user input. The present invention may then include translating the received user input into an embedding matrix and inputting the embedding matrix into a deep neural network. The present invention may further include generating, by the deep neural network, an output vector, a user bias term and an item bias term based on the embedding matrix. The present invention may then include calculating a predicted rating based on the generated output vector, the generated user bias term and the generated item bias term. The present invention may then include determining an accuracy of the predicted rating.