Abstract: Methods of fabricating a probe are described. In an example, a structure may be formed on a surface of a substrate. The structure may include the probe, a hinge, and an anchor arranged linearly, where an angle is formed between the probe and the hinge. The hinge may be positioned between the probe and the anchor, and the structure may be parallel to the substrate. An amount of solder may be deposited on an area of the structure that spans from a portion of the probe to a portion of the anchor, and across the hinge. The deposited solder may be reshaped by an execution of a solder reflow process. The reshape of the deposited solder may cause the probe to rotate about the hinge in order to reduce the angle between the probe and the hinge.

Abstract: An apparatus includes a memory and a processor coupled to the memory. The processor includes first and second sets of arithmetic units having first and second precision for floating-point computations, the second precision being lower than the first precision. The processor is configured to obtain a machine learning model trained in the first precision, to utilize the second set of arithmetic units to perform inference on input data, to utilize the first set of arithmetic units to generate feedback for updating parameters of the second set of arithmetic units based on the inference performed on the input data by the second set of arithmetic units, to tune parameters of the second set of arithmetic units based at least in part on the feedback generated by the first set of arithmetic units, and to utilize the second set of arithmetic units with the tuned parameters to generate inference results.

Abstract: An apparatus for training and inferencing a neural network includes circuitry that is configured to generate a first weight having a first format including a first number of bits based at least in part on a second weight having a second format including a second number of bits and a residual having a third format including a third number of bits. The second number of bits and the third number of bits are each less than the first number of bits. The circuitry is further configured to update the second weight based at least in part on the first weight and to update the residual based at least in part on the updated second weight and the first weight. The circuitry is further configured to update the first weight based at least in part on the updated second weight and the updated residual.

Abstract: An apparatus includes circuitry for a neural network that is configured to perform forward propagation neural network operations on floating point numbers having a first n-bit floating point format. The first n-bit floating point format has a configuration consisting of a sign bit, m exponent bits and p mantissa bits where m is greater than p. The circuitry is further configured to perform backward propagation neural network operations on floating point numbers having a second n-bit floating point format that is different than the first n-bit floating point format. The second n-bit floating point format has a configuration consisting of a sign bit, q exponent bits and r mantissa bits where q is greater than m and r is less than p.

Abstract: A semiconductor device that includes at least one germanium containing fin structure having a length along a <100> direction and a sidewall orientated along the (100) plane. The semiconductor device also includes at least one germanium free fin structure having a length along a <100> direction and a sidewall orientated along the (100) plane. A gate structure is present on a channel region of each of the germanium containing fin structure and the germanium free fin structure. N-type epitaxial semiconductor material having a square geometry present on the source and drain portions of the sidewalls having the (100) plane orientation of the germanium free fin structures. P-type epitaxial semiconductor material having a square geometry is present on the source and drain portions of the sidewalls having the (100) plane orientation of the germanium containing fin structures.

Abstract: Embodiments for implementing a communicating memory between a plurality of computing components are provided. In one embodiment, an apparatus comprises a plurality of memory components residing on a processing chip, the plurality of memory components interconnected between a plurality of processing elements of at least one processing core of the processing chip and at least one external memory component external to the processing chip. The apparatus further comprises a plurality of load agents and a plurality of store agents on the processing chip, each interfacing with the plurality of memory components. Each of the plurality of load agents and the plurality of store agents execute an independent program specifying a destination of data transacted between the plurality of memory components, the at least one external memory component, and the plurality of processing elements.

Abstract: A semiconductor device that includes at least one germanium containing fin structure having a length along a <100> direction and a sidewall orientated along the (100) plane. The semiconductor device also includes at least one germanium free fin structure having a length along a <100> direction and a sidewall orientated along the (100) plane. A gate structure is present on a channel region of each of the germanium containing fin structure and the germanium free fin structure. N-type epitaxial semiconductor material having a square geometry present on the source and drain portions of the sidewalls having the (100) plane orientation of the germanium free fin structures. P-type epitaxial semiconductor material having a square geometry is present on the source and drain portions of the sidewalls having the (100) plane orientation of the germanium containing fin structures.

Abstract: Embodiments for implementing a communicating memory between a plurality of computing components are provided. In one embodiment, an apparatus comprises a plurality of memory components residing on a processing chip, the plurality of memory components interconnected between a plurality of processing elements of at least one processing core of the processing chip and at least one external memory component external to the processing chip. The apparatus further comprises a plurality of load agents and a plurality of store agents on the processing chip, each interfacing with the plurality of memory components. Each of the plurality of load agents and the plurality of store agents execute an independent program specifying a destination of data transacted between the plurality of memory components, the at least one external memory component, and the plurality of processing elements.

Abstract: Techniques facilitating matrix multiplication on a systolic array are provided. A computer-implemented method can comprise populating, by a system operatively coupled to a processor, respective first registers of one or more processing elements of a systolic array structure with respective input data bits of a first data matrix. The one or more processing elements can comprise a first processing element that comprises a first input data bit of the first data matrix and a first activation bit of a second data matrix. The method can also include determining, by the system, at the first processing element, a first partial sum of a third data matrix. Further, the method can include streaming, by the system, the first partial sum of the third data matrix from the first processing element.

Abstract: A semiconductor device that includes at least one germanium containing fin structure having a length along a <100> direction and a sidewall orientated along the (100) plane. The semiconductor device also includes at least one germanium free fin structure having a length along a <100> direction and a sidewall orientated along the (100) plane. A gate structure is present on a channel region of each of the germanium containing fin structure and the germanium free fin structure. N-type epitaxial semiconductor material having a square geometry present on the source and drain portions of the sidewalls having the (100) plane orientation of the germanium free fin structures. P-type epitaxial semiconductor material having a square geometry is present on the source and drain portions of the sidewalls having the (100) plane orientation of the germanium containing fin structures.

Abstract: An adversarial robustness testing method, system, and computer program product include testing a robustness of a black-box system under different access settings via an accelerator.

Abstract: A processor receives input data and provides the input data to a first neural network including a first neural network model. The first neural network model has a first numerical precision level. A first feature vector is generated from the input data using the first neural network. The input data is provided to a second neural network including a second neural network model. The second neural network model has a second numerical precision level different from the first numerical precession level. A second feature vector is generated from the input data using the second neural network. A difference metric is computed between the first feature vector and the second feature vector. The difference metric is indicative of whether the input data includes adversarial data.

Abstract: Methods of fabricating a probe are described. In an example, a structure may be formed on a surface of a substrate. The structure may include the probe, a hinge, and an anchor arranged linearly, where an angle is formed between the probe and the hinge. The hinge may be positioned between the probe and the anchor, and the structure may be parallel to the substrate. An amount of solder may be deposited on an area of the structure that spans from a portion of the probe to a portion of the anchor, and across the hinge. The deposited solder may be reshaped by an execution of a solder reflow process. The reshape of the deposited solder may cause the probe to rotate about the hinge in order to reduce the angle between the probe and the hinge.

Abstract: Systems, computer-implemented methods, and computer program products to facilitate gradient weight compression are provided. According to an embodiment, a system can comprise a memory that stores computer executable components and a processor that executes the computer executable components stored in the memory. The computer executable components can comprise a pointer component that can identify one or more compressed gradient weights not present in a first concatenated compressed gradient weight. The computer executable components can further comprise a compression component that can compute a second concatenated compressed gradient weight based on the one or more compressed gradient weights to update a weight of a learning entity of a machine learning system.

Abstract: A method for forming a hybrid complementary metal oxide semiconductor (CMOS) device includes orienting a semiconductor layer of a semiconductor-on-insulator (SOI) substrate with a base substrate of the SOI, exposing the base substrate in an N-well region by etching through a mask layer, a dielectric layer, the semiconductor layer and a buried dielectric to form a trench and forming spacers on sidewalls of the trench. The base substrate is epitaxially grown from a bottom of the trench to form an extended region. A fin material is epitaxially grown from the extended region within the trench. The mask layer and the dielectric layer are restored over the trench. P-type field-effect transistor (PFET) fins are etched on the base substrate, and N-type field-effect transistor (NFET) fins are etched in the semiconductor layer.

Abstract: A Static Random Access Memory (SRAM) device in a binary neural network is provided. The SRAM device includes an SRAM inference engine having an SRAM computation architecture with a forward path that include multiple SRAM cells. The multiple SRAM cells are configured to form a chain of SRAM cells such that an output of a given one of the multiple SRAM cells is an input to a following one of the multiple SRAM cells. The SRAM computation architecture is configured to compute a prediction from an input.

Abstract: A method for improving performance of a predefined Deep Neural Network (DNN) convolution processing on a computing device includes inputting parameters, as input data into a processor on a computer that formalizes a design space exploration of a convolution mapping, on a predefined computer architecture that will execute the predefined convolution processing. The parameters are predefined as guided by a specification for the predefined convolution processing to be implemented by the convolution mapping and by a microarchitectural specification for the processor that will execute the predefined convolution processing. The processor calculates performance metrics for executing the predefined convolution processing on the computing device, as functions of the predefined parameters, as proxy estimates of performance of different possible design choices to implement the predefined convolution processing.

Abstract: A semiconductor device that includes at least one germanium containing fin structure having a length along a <100> direction and a sidewall orientated along the (100) plane. The semiconductor device also includes at least one germanium free fin structure having a length along a <100> direction and a sidewall orientated along the (100) plane. A gate structure is present on a channel region of each of the germanium containing fin structure and the germanium free fin structure. N-type epitaxial semiconductor material having a square geometry present on the source and drain portions of the sidewalls having the (100) plane orientation of the germanium free fin structures. P-type epitaxial semiconductor material having a square geometry is present on the source and drain portions of the sidewalls having the (100) plane orientation of the germanium containing fin structures.

Abstract: A random number signal generator used for performing dropout or weight initialization for a node in a neural network. The random number signal generator includes a transistor which generates a random noise signal. The transistor includes a substrate, source and drain regions formed in the substrate, a first insulating layer formed over a channel of the transistor, a first trapping layer formed over the first insulating layer, a second insulating layer formed over the first trapping layer, and a second trapping layer formed over the second insulating layer. One or more traps in the first and second trapping layers are configured to capture or release one or more carriers flowing through the channel region. The random noise signal is generated as a function of one or more carrier being captured or released by the one or more traps.

Abstract: A computer-implemented method, computer program product, and system for determination of critical parts and component correlations in a circuit using a correlation graph and centrality analysis including; receiving a circuit layout portion of a larger circuit layout, converting the circuit layout portion into a correlation graph representing components as nodes and connecting wires as edges, determining, using ground truth and Naïve Bayes to determine correlation weighting, scaling the correlation graph to represent the larger circuit, and presenting the larger correlation graph on a graphical user interface (GUI).