Abstract: An example apparatus having a heterogenous memory system includes a first sensor layer, of a plurality of stacked sensor layers, including an array of pixels; and one or more semiconductor layers of the plurality of stacked sensor layers located beneath the first sensor layer, the one or more semiconductor layers configured to process pixel data output by the array of pixels, the one or more semiconductor layers including a first memory to store most significant bits (“MSBs”) of data involved in the processing of the pixel data; a second memory to store least significant bits (“LSBs”) of the data; and wherein the first memory has a lower bit error rate (“BER”) than the second memory.

Abstract: A console and headset system locally trains machine learning models to perform customized online learning tasks. To customize the online learning models for specific users of the system without using outside resources, the system trains the models to compare a target frame to stored calibration frames, rather than directly inferring information about a target frame. During deployment, an embedding is generated for the target frame. A sample embedding that is closest to the target embedding is selected from a group of embeddings of calibration frames. The information about the selected embedding and target embedding and ground truths for the calibration frame are provided as inputs to one of the trained models. The model predicts a difference between the target frame and the calibration frame, which can be used to determine information about the target frame.

Abstract: Methods, systems, and media for partitioning neural networks are provided. In some embodiments, a method comprises obtaining a training set. The method comprises training a plurality of neural networks using the training set, wherein neural networks differ based on dimensions of one or more layers of the neural networks and a location of a compression block positioned between a first set of layers of a neural network and a second set of layers of the neural network. The method comprises selecting a neural network based on hardware constraints of a system on which the neural network is to be implemented, wherein the first set of layers of the selected neural network are executed by one or more sensor devices of the system, and wherein the second set of layers of the selected neural network are executed by an aggregator computing device of the system.

Abstract: The disclosed computer-implemented method may include (i) conditionally operating, at a first frequency, a first stage of an eye-tracking system processing pipeline that detects a region of interest and (ii) operating, at a second frequency that is substantially greater than the first frequency, a second stage of the eye-tracking system processing pipeline that predicts a gaze orientation based at least in part on the detected region of interest. Various other methods, systems, and computer-readable media are also disclosed.

Abstract: The disclosed system may include a first layer that includes multiple digital pixel sensors configured to detect light. The system may also include a second layer that includes various image processing components configured to process the light detected by the digital pixel sensors. Still further, the system may include a third layer that includes machine learning (ML) hardware processing components. The image processing components of the second layer may be communicatively connected to the ML hardware processing components of the third layer via multiple micro through-silicon vias (uTSVs). Various other methods of manufacturing, apparatuses, and computer-readable media are also disclosed.

Abstract: In one example, an apparatus comprises: a first sensor layer, including an array of pixel cells configured to generate pixel data; and one or more semiconductor layers located beneath the first sensor layer with the one or more semiconductor layers being electrically connected to the first sensor layer via interconnects. The one or more semiconductor layers comprises on-chip compute circuits configured to receive the pixel data via the interconnects and process the pixel data, the on-chip compute circuits comprising: a machine learning (ML) model accelerator configured to implement a convolutional neural network (CNN) model to process the pixel data; a first memory to store coefficients of the CNN model and instruction codes; a second memory to store the pixel data of a frame; and a controller configured to execute the codes to control operations of the ML model accelerator, the first memory, and the second memory.

Abstract: A stacked camera-image-sensor circuit may include (i) a first layer that includes a plurality of image sensing elements, (ii) a second layer that includes components that interface with the image sensing elements, and (iii) at least one additional layer that includes image-processing components. Various other methods, systems, and computer-readable media are also disclosed.

Abstract: A sensor assembly for determining one or more features of a local area is presented herein. The sensor assembly includes a plurality of stacked sensor layers. A first sensor layer of the plurality of stacked sensor layers located on top of the sensor assembly includes an array of pixels. The top sensor layer can be configured to capture one or more images of light reflected from one or more objects in the local area. The sensor assembly further includes one or more sensor layers located beneath the top sensor layer. The one or more sensor layers can be configured to process data related to the captured one or more images. Different sensor architectures featuring various arrangements of memory and computing devices are described, some of which feature in-memory computing. A plurality of sensor assemblies can be integrated into an artificial reality system, e.g., a head-mounted display.

Abstract: A neuromorphic system comprises a set of at least one input neuron, a set of at least one output neuron and a synaptic network formed from a set of at least one variable-resistance memristive component, said synaptic network connecting at least one input neuron to at least one output neuron, the resistance of the at least one memristive component being adjusted by delivering to the synaptic network write pulses generated by the at least one input neuron, and return pulses generated by the at least one output neuron, the characteristics of the write and return pulses being deduced from the intrinsic characteristics of the at least one memristive component so that the combination of a write pulse and a return pulse in the at least one memristive component results in a modification of its resistance according to a learning rule chosen beforehand.

Abstract: A microelectronic flash memory device including a plurality of memory cells including transistors fitted with a matrix of channels connecting a block of common source to a second block on which bit lines rest, the transistors also being formed by a plurality of gates including at least one gate material, including a first selection gate coating the channels, a plurality of control gates coating the channels, a plurality of second selection gates each coating the channels of the same row and the matricial arrangement, at least one or more of the gates based on superposition of layers including at least one first layer of dielectrical material, at least one charge store zone, and at least one second layer of dielectrical material.

Abstract: A delay generator comprises at least one programmable resistor RPCM made of a chalcogenide-based phase-change material, said resistor RPCM being initialized, so as to generate a delay, in a way such that the resistance of the resistor RPCM equals a pre-set initial value R0 and such that the chalcogenide is in the amorphous phase, and a comparator comparing a reference electrical quantity that is stable over time with a variable electrical quantity representative of the resistance of the programmable resistor RPCM, the comparator generating a singularity signal s, said singularity being generated when the difference between the two electrical quantities changes sign.

Abstract: A neuromorphic system comprises a set of at least one input neuron, a set of at least one output neuron and a synaptic network formed from a set of at least one variable-resistance memristive component, said synaptic network connecting at least one input neuron to at least one output neuron, the resistance of the at least one memristive component being adjusted by delivering to the synaptic network write pulses generated by the at least one input neuron, and return pulses generated by the at least one output neuron, the characteristics of the write and return pulses being deduced from the intrinsic characteristics of the at least one memristive component so that the combination of a write pulse and a return pulse in the at least one memristive component results in a modification of its resistance according to a learning rule chosen beforehand.

Abstract: A process for fabricating a charge storage layer comprising metal particles of a memory cell, said layer consisting of an organic layer comprising, on the surface, said metal particles, said process comprising the following steps: (a) a step of grafting, onto a metallic, semiconductor or electrically insulating substrate, an organic layer comprising, on the surface, groups capable of complexing at least one metallic element in cationic form; (b) a step of bringing said layer into contact with a solution comprising said metallic element in cationic form, by means of which said metallic element is complexed by said abovementioned groups; and (c) a step of reducing said complexed metallic element to the metallic element in oxidation state 0, by means of which metal particles are obtained.

Abstract: A delay generator comprises at least one programmable resistor RPCM made of a chalcogenide-based phase-change material, said resistor RPCM being initialized, so as to generate a delay, in a way such that the resistance of the resistor RPCM equals a pre-set initial value R0 and such that the chalcogenide is in the amorphous phase, and a comparator comparing a reference electrical quantity that is stable over time with a variable electrical quantity representative of the resistance of the programmable resistor RPCM, the comparator generating a singularity signal s, said singularity being generated when the difference between the two electrical quantities changes sign.

Abstract: A microelectronic flash memory device including a plurality of memory cells including transistors fitted with a matrix of channels connecting a block of common source to a second block on which bit lines rest, the transistors also being formed by a plurality of gates including at least one gate material, including a first selection gate coating the channels, a plurality of control gates coating the channels, a plurality of second selection gates each coating the channels of the same row and the matricial arrangement, at least one or more of the gates based on superposition of layers including at least one first layer of dielectrical material, at least one charge store zone, and at least one second layer of dielectrical material.

Abstract: A process for fabricating a charge storage layer comprising metal particles of a memory cell, said layer consisting of an organic layer comprising, on the surface, said metal particles, said process comprising the following steps: (a) a step of grafting, onto a metallic, semiconductor or electrically insulating substrate, an organic layer comprising, on the surface, groups capable of complexing at least one metallic element in cationic form; (b) a step of bringing said layer into contact with a solution comprising said metallic element in cationic form, by means of which said metallic element is complexed by said abovementioned groups; and (c) a step of reducing said complexed metallic element to the metallic element in oxidation state 0, by means of which metal particles are obtained.