Abstract: Embodiments are generally directed to dynamically dividing activations and kernels for improving memory efficiency. An embodiment of a method in a compute engine performing machine learning comprises: receiving, by a convolutional layer of a convolutional neural network (CNN) implemented on the compute engine, a plurality of activation groups contained in an input data, wherein the convolutional layer includes one or more kernel groups and the one or more kernel groups each include a plurality of kernels; determining a plurality of memory efficiency metrics based on the number of activation groups of the plurality of activation groups and the number of kernels of the plurality of kernels; selecting a first optimal number of activation groups and a second optimal number of kernels that are associated with an optimal memory efficiency metric in the plurality of memory efficiency metrics; and performing a convolutional operation on the input data based on the first optimal number and the second optimal number.

Abstract: A connector assembly includes a connector and a locking structure. The connector includes an insulating body. The insulating body includes at least one mating end portion. The mating end portion defines a first mating space for receiving a conductive terminal. The locking structure is held to the mating end portion. The locking structure has a pre-locking position and a locking position relative to the mating end portion. The locking structure includes a body portion and a resisting portion. The body portion is configured to receive a high voltage interlock terminal. The resisting portion is configured such that when the locking structure is at the pre-tightened position, the resisting portion is located outside the first mating space; when the locking structure is at the locking position, at least part of the resisting portion protrudes into the first mating space.

Abstract: One embodiment provides a graphics processor comprising an instruction cache to store an instruction and a compute block configured to perform multiply-accumulate operations in response to execution of the instruction. The compute block includes a scheduler to schedule a plurality of threads for execution of the instruction and multiply-accumulate circuitry configured to execute the instruction via the plurality of threads, wherein the multiply-accumulate circuitry includes a plurality of functional units configured to process, in parallel via the plurality of threads, a corresponding plurality of matrix elements to multiply a first matrix and a second matrix, and to multiply the first matrix and the second matrix includes to multiply data elements in a row of the first matrix by corresponding data elements in a column of the second matrix to generate a plurality of products.

Abstract: The present disclosure provides aqueous formulations comprising hydrocortisone sodium phosphate and monothioglycerol. In some embodiments, the formulations comprise monobasic sodium phosphate, dibasic sodium phosphate, or disodium EDTA. The present disclosure further provides a method of treating a disease or disorder in a subject by administering the aqueous formulation.

Abstract: One embodiment provides for a compute apparatus comprising a decode unit to decode a single instruction into a decoded instruction that specifies multiple operands including a multi-bit input value and a one-bit weight associated with a neural network, as well as an arithmetic logic unit including a multiplier, an adder, and an accumulator register. To execute the decoded instruction, the multiplier is to perform a fused operation including an exclusive not OR (XNOR) operation and a population count operation. The adder is configured to add the intermediate product to a value stored in the accumulator register and update the value stored in the accumulator register.

Abstract: Some embodiments pertain to generating, transmitting, and using ionospheric disturbance information applicable to at least a part of the Earth's surface. The ionospheric disturbance information is transmitted in the form of at least one message comprising reference point ionospheric disturbance levels, i.e. levels associated with a plurality of reference points on a reference ionospheric shell. On the receiver side, an estimator is operated, and, for each of at least one NSS signal observed by a NSS receiver, the reference point ionospheric disturbance levels are used to decide whether to adopt some ionospheric disturbance mitigation measures in the context of estimating parameters useful to determine a position. The ionospheric disturbance information may for example comprise ionospheric amplitude scintillation information, ionospheric phase scintillation information, and/or ionospheric gradient information. Systems, vehicles, and computer programs are also disclosed.

Abstract: An integrated circuit (IC) package apparatus is disclosed. The IC package includes one or more processing units and a bridge, mounted below the one or more processing unit, including one or more arithmetic logic units (ALUs) to perform atomic operations.

Abstract: The present disclosure provides aqueous formulations comprising hydrocortisone sodium phosphate and monothioglycerol. In some embodiments, the formulations comprise monobasic sodium phosphate, dibasic sodium phosphate, or disodium EDTA. The present disclosure further provides a method of treating a disease or disorder in a subject by administering the aqueous formulation.

Abstract: The invention relates to methods for providing geodetic datum transformation information applicable to a region of the Earth, for use by a navigation satellite system (NSS) receiver and/or a processing entity capable of receiving data therefrom. In one embodiment, the method comprises: generating (s10a) similarity transformation parameters and, for each of a plurality of reference stations in the region, post-fit transformation residuals applicable to the reference station; and sending (s20a) the transformation parameters and output residuals applicable to output points, wherein an output residual is one of the post-fit transformation residuals or derived therefrom; and the distribution of the output points in the region depends on the distribution of the reference stations.

Abstract: This disclosure provides a resource tapping method, a resource tapping apparatus and an electronic device, and relates to the field of computer technology, in particular to the technical field of artificial intelligence, such as deep learning and machine learning. A specific implementation is as follows: obtaining operation data in M resource dimensions of a target cabinet, the M resource dimensions including a power resource, where M is a positive integer; determining a target power over-allocation value of the target cabinet based on the operation data, the target power over-allocation value being used for indicating an allowable power increment on the basis of a power rating of the target cabinet; and determining, based on the target power over-allocation value, a first quantity of additional servers deployable in the target cabinet.

Abstract: The present disclosure relates to formulations of Compound A: or a pharmaceutically acceptable salt thereof. The present disclosure also relates to methods of manufacturing such formulations and uses of those formulations in treating prostate cancer.

Abstract: A conductive terminal includes a first fixing portion, a first mating portion, a second fixing portion and a second mating portion. The second mating portion and the first mating portion are arranged side by side along a first direction. The first mating portion includes a first elastic piece extending along a second direction. The first elastic piece includes a first elastic contact portion. The second mating portion includes a second elastic piece. The second elastic piece includes a second elastic contact portion. The first elastic contact portion extends along a first arc-shaped surface. The second elastic contact portion extends along a second arc-shaped surface. Protruding directions of the first arc-shaped surface and the second arc-shaped surface are opposite to each other along the first direction. An electrical connector and an electrical connector assembly having the conductive terminal are disclosed.

Abstract: One embodiment provides for a compute apparatus comprising a decode unit to decode a single instruction into a decoded instruction that specifies multiple operands including a multi-bit input value and a one-bit weight associated with a neural network, as well as an arithmetic logic unit including a multiplier, an adder, and an accumulator register. To execute the decoded instruction, the multiplier is to perform a fused operation including an exclusive not OR (XNOR) operation and a population count operation. The adder is configured to add the intermediate product to a value stored in the accumulator register and update the value stored in the accumulator register.

Abstract: In an example, an apparatus comprises a compute engine comprising a high precision component and a low precision component; and logic, at least partially including hardware logic, to receive instructions in the compute engine; select at least one of the high precision component or the low precision component to execute the instructions; and apply a gate to at least one of the high precision component or the low precision component to execute the instructions. Other embodiments are also disclosed and claimed.

Abstract: Described herein is a general-purpose graphics processing unit including a multiprocessor having a single instruction, multiple thread, SIMT, architecture. The multiprocessor comprises multiple sets of compute units each having a first logic unit configured to perform floating-point operations and a second logic unit configured to perform integer operations, with a thread of the floating-point instruction being executed in parallel with a thread of the integer instruction.

Abstract: This application relates to a roller device. The device comprises: a roller; and a roller axle, the roller axle comprises: an axle portion configured to mount the roller such that the roller is rotatable about the axle portion; and a riveting portion at a mounting end of the roller axle with respect to an axial direction of the roller axle, and extending from the mounting end in a direction away from the axle portion, wherein, when the roller device is pressed against a carrier under force, the riveting portion can pierce the carrier such that the roller axle is riveted with the carrier through the riveting portion.

Abstract: Disclosed are a coordinated control method and system for active support of a new energy power station containing a voltage-controlled source. The method includes: obtaining a working mode and a controlled target value of an external characteristic of a new energy power station, and setting internal parameters of a voltage-controlled source and a current-controlled source in the new energy power station; monitoring a power output state of a controlled unit of a voltage source, and calculating a regulation margin of the voltage source; setting power, inertia, and an objective function and a constraint of multi-source coordinated control of primary frequency modulation (FM)/voltage regulation for the new energy power station, and obtaining controlled allocation logic of a current source based on the objective function and the constraint of the multi-source coordinated control; and setting secondary control logic of each controlled unit and its allocation priority strategy.

Abstract: One embodiment provides a general-purpose graphics processing unit comprising a dynamic precision floating-point unit including a control unit having precision tracking hardware logic to track an available number of bits of precision for computed data relative to a target precision, wherein the dynamic precision floating-point unit includes computational logic to output data at multiple precisions.