Abstract: Various implementations described herein are directed to a device having a first read wordline formed in a first metal layer and a read wordline driver having an output node coupled to one or more memory cells via the first read wordline formed in the first metal layer. The device may include a second read wordline formed in a second metal layer that is different than the first metal layer, and the read wordline driver may have an input node coupled to the second read wordline formed in the second metal layer.

Abstract: Techniques are disclosed for using a trained machine learning (ML) pipeline to identify categories associated with target data items even though the identified categories may not already be present in the hierarchy. The ML pipeline may include trained cluster-based and classification-based machine learning models, among others. If the results of the cluster-based and classification-based machine learning models are the same, then the target data items is assigned to a hierarchical classification consistent with the identical results of the machine learning model. An assigned hierarchical classification may be validated by the operation of subsequent trained ML models that determine whether parent and child categories in the identified classification are properly associated with one another.

Abstract: Various implementations described herein are related to a device with a wordline driver that provides a wordline signal to a wordline based on a row selection signal and a row clock signal. The device may have row selector logic that provides the row selection signal to the wordline driver based on first input signals in a periphery voltage domain. The device may also have level shifter circuitry that provides the row clock signal to the wordline driver in a core voltage domain based on second input signals in the periphery voltage domain.

Abstract: Methods and systems for revoking electronic messages. One method includes storing, for each of a plurality of forwarded messages, a record in a data store, each record including a link to an original message for the forwarded message, and receiving a request to revoke a forwarded message. In response to receiving the request, the method includes identifying an original message the forwarded message via a record stored in the data store and notifying, with an electronic processor, a user associated with the original message of the request to revoke the forwarded message. In response to receiving an instruction revoking the original source message from the user, the method includes identifying each forward of the original message via records stored in the data store and revoking the original message and each message associated with each record stored in the data store including a link to the original message.

Abstract: A system according to an embodiment optimizes communications with users using machine learning based models. The system receives user profile data for a set of users. For each user from the set of users, the system provides the user profile data as input to a machine learning based model and determines attributes describing the user, for example a measure of adherence rate for the user. The system ranks the set of users based on the predicted attributes. The system selects a subset of users from the set of users based on the ranking. For each selected user from the set of selected users, the system determines communication parameters for communicating with the selected user and sends a communication to the selected user based on the determined communication parameters.

Abstract: Various implementations described herein are related to a device with a wordline driver that provides a wordline signal to a wordline based on a row selection signal and a row clock signal. The device may have row selector logic that provides the row selection signal to the wordline driver based on first input signals in a periphery voltage domain. The device may also have level shifter circuitry that provides the row clock signal to the wordline driver in a core voltage domain based on second input signals in the periphery voltage domain.

Abstract: Various implementations described herein are related to a method for accessing a bitcell in an array of bitcells with a wordline and a bitline. The method may perform a precharge operation on the bitline that precharges the bitline after a read cycle and before a write cycle. Also, the method may extend precharge time of the precharge operation between the read cycle and the write cycle, e.g., by modulating a wordline signal on the wordline with early cut-off of the wordline signal on the wordline during the read cycle.

Abstract: Various implementations described herein are directed to a device having a first read wordline formed in a first metal layer and a read wordline driver having an output node coupled to one or more memory cells via the first read wordline formed in the first metal layer. The device may include a second read wordline formed in a second metal layer that is different than the first metal layer, and the read wordline driver may have an input node coupled to the second read wordline formed in the second metal layer.

Abstract: Embodiments relate to configuring artificial-intelligence (AI) decision nodes throughout a communication decision tree. The decision nodes can support successive iteration of AI models to dynamically define iteration data that corresponds to a trajectory through the tree.

Abstract: Techniques are disclosed for training a machine learning model to identify and rank entities relative to a set of requirements. The trained machine learning model may present an array of interface elements (e.g., icons) in a graphical user interface (GUI), where the interface elements represent corresponding entities. These interface elements are arranged in the GUI based on their corresponding ranks. The ranks of entities, and therefore the locations of corresponding interface elements are based, at least in part, on a degree of match between values of a subset of entity attributes and a corresponding subset of the set of requirements. The machine learning model may be further trained by receiving a user input that changes a location of a particular user interface element within the graphical user interface displaying the ranked user interface elements.

Abstract: Techniques are disclosed for using a machine learning model to identify and present a ranked array of interface elements representing entities. The location of individual interface elements within the ranked array of interface elements is based on a level of match between entity attributes and a set of requirements established by a user. The machine learning model may be further trained by receiving a user input that changes a location of a particular user interface element within a graphical user interface displaying the ranked array. Upon receiving the user input, the trained machine learning model may update training data to include an updated match score for the particular user interface element that reflects the new location.

Abstract: Various implementations described herein are related to a device having memory circuitry with bitlines coupled to an array of bitcells. Also, the device may have first precharge circuitry that precharges the bitlines before a write cycle. Also, the device may have second precharge circuitry that precharges the bitlines after the write cycle.

Abstract: An integrated circuit device includes a programmable fabric that has a plurality of memory blocks. The integrated circuit device also includes a network-on-chip (NOC) located on a shoreline of the programmable fabric and at least one micro NOC formed with hardened resources in the programmable fabric. The at least one micro NOC is communicatively coupled to the NOC and to at least one memory block of the plurality of memory blocks. Additionally, the at least one micro NOC is configurable to route data between the NOC and the at least one memory block.

Abstract: Various implementations described herein are related to a device having multiple transistors that are arranged as a bitcell. The device may include multiple wordlines that are coupled to the multiple transistors. Also, one or more wordlines may be formed with frontside metal, and one or more other wordlines may be formed with buried metal.

Abstract: The invention provides a method and system for operating an inverter based distributed power generation source with energy storage system, as a Flexible AC Transmission System (FACTS) device—a STATCOM. The inverter based distributed power generation source can provide reactive power compensation, voltage regulation, damping enhancement, stability improvement and other benefits provided by FACTS devices. These STATCOM functions are provided when the said energy storage based distributed power generation source is doing at least one of: i) not exchanging active power with said power grid system, or ii) exchanging active power less than a maximum inverter capacity with said power grid system. The present invention thus provides a technological improvement that opens up a new set of applications and potential revenue earning opportunities for energy storage based distributed power generation sources other than simply from exchanging (injecting or absorbing) active power.

Abstract: Systems, methods, and devices relating to operating a power generation facility to contribute to the stability of the power transmission system. A controller operates on the power generation facility to modulate real power or reactive power or both in a decoupled manner to contribute to the stability of the power transmission system. Real power produced by the power generation facility can be increased or decreased between zero and the maximum real power available from the PV solar panels, as required by the power system. Reactive power from the power generation facility can be exchanged (injected or absorbed) and both increased or decreased as required by the power transmission system. For solar farms, the solar panels can be connected or disconnected, or operated at non-optimal power production to add or subtract real or reactive power to the power transmission system.

Abstract: Embodiments relate to configuring artificial-intelligence (AI) decision nodes throughout a communication decision tree. The decision nodes can support successive iteration of AI models to dynamically define iteration data that corresponds to a trajectory through the tree.

Abstract: Various implementations described herein are related to a device having wordline drivers coupled to a core array. The device may have backside power network with buried power rails. The device may have header logic coupled to power supply connections of the wordline drivers by way of the buried power rails, and the header logic may be used to power-gate the wordline drivers.

Abstract: Various implementations described herein are directed to a device having memory with a first array and a second array. The device may have power rails formed in frontside metal layers that supply core voltage to the memory. The power rails may include a first path routed through a first frontside metal layer to the first array of the memory, and the power rails may include a second path routed through the first frontside metal layer and a second frontside metal layer to the second array of the memory.

Abstract: Various implementations described herein are related to a device having logic that operates in multiple voltage domains. The device may include a backside power network with rows of segmented supply rails coupled to the logic. The rows of segmented supply rails may include alternating rail breaks that define an interchanging directional supply of power to the logic in the multiple voltage domains.