Samsung Electronics Co., Ltd.

Borna Josip Obradovic

Mark Stephen Rodder

Titash Rakshit

Jorge A Kittl

Ryan M Hatcher

Joon Goo Hong

Robert Christopher Bowen

Wei-E Wang

Rwik Sengupta

Dharmendar Reddy Palle

Chris Bowen

Ganesh Hegde

Kang Ill Seo

Seung Hun Lee

Vladimir Nikitin

Dmytro Apalkov

Kang-ill Seo

Engin Ipek

Pan Kwi Park

Seung Ryul Lee

Palle Dharmendar

Mohammad Ali Pourghaderi

A neuromorphic device for the analog computation of a linear combination of input signals, for use, for example, in an artificial neuron. The neuromorphic device provides non-volatile programming of the weights, and fast evaluation and programming, and is suitable for fabrication at high density as part of a plurality of neuromorphic devices. The neuromorphic device is implemented as a vertical stack of flash-like cells with a common control gate contact and individually contacted source-drain (SD) regions. The vertical stacking of the cells enables efficient use of layout resources.

High-density neuromorphic computing element

A neuromorphic weight cell (NWC) including a resistor ladder including a plurality of resistors connected in series, and a plurality of shunting nonvolatile memory (NVM) elements, each of the shunting NVM elements being coupled in parallel to a corresponding one of the resistors.

Monolithic multi-bit weight cell for neuromorphic computing

A method of storing a sparse weight matrix for a trained artificial neural network in a circuit including a series of clusters. The method includes partitioning the sparse weight matrix into at least one first sub-block and at least one second sub-block. The first sub-block includes only zero-value weights and the second sub-block includes non-zero value weights. The method also includes assigning the non-zero value weights in the at least one second sub-block to at least one cluster of the series of clusters of the circuit. The circuit is configured to perform matrix-vector-multiplication (MVM) between the non-zero value weights of the at least one second sub-block and an input vector during an inference process utilizing the artificial neural network. The sub-blocks containing all zero elements are power gated, thereby reducing overall energy consumption for inference.

Method of enabling sparse neural networks on memresistive accelerators

A neuromorphic multi-bit digital weight cell configured to store a series of potential weights for a neuron in an artificial neural network. The neuromorphic multi-bit digital weight cell includes a parallel cell including a series of passive resistors in parallel and a series of gating transistors. Each gating transistor of the series of gating transistors is in series with one passive resistor of the series of passive resistors. The neuromorphic cell also includes a series of programming input lines connected to the series of gating transistors, an input terminal connected to the parallel cell, and an output terminal connected to the parallel cell.

Multi-bit, SoC-compatible neuromorphic weight cell using ferroelectric FETs

A method and system for training a neural network are described. The method includes providing at least one continuously differentiable model of the neural network. The at least one continuously differentiable model is specific to hardware of the neural network. The method also includes iteratively training the neural network using the at least one continuously differentiable model to provide at least one output for the neural network. Each iteration uses at least one output of a previous iteration and a current continuously differentiable model of the at least one continuously differentiable model.

Method and system for training of neural networks using continuously differentiable models

A method provides a gate structure for a plurality of components of a semiconductor device. A silicate layer is provided. In one aspect, the silicate layer is provided on a channel of a CMOS device. A high dielectric constant layer is provided on the silicate layer. The method also includes providing a work function metal layer on the high dielectric constant layer. A low temperature anneal is performed after the high dielectric constant layer is provided. A contact metal layer is provided on the work function metal layer.

Method of forming multi-threshold voltage devices and devices so formed

A semiconductor device includes a series of metal routing layers and a complementary pair of planar field-effect transistors (FETs) on an upper metal routing layer of the metal routing layers. The upper metal routing layer is M3 or higher. Each of the FETs includes a channel region of a crystalline material. The crystalline material may include polycrystalline silicon. The upper metal routing layer M3 or higher may include cobalt.

Semiconductor device including a repeater/buffer at upper metal routing layers and methods of manufacturing the same

A hardware cell and method for performing a digital XNOR of an input signal and weights are described. The hardware cell includes input lines, a plurality of pairs of magnetic junctions, output transistors and at least one selection transistor coupled with the output transistors. The input lines receive the input signal and its complement. The magnetic junctions store the weight. Each magnetic junction includes a reference layer, a free layer and a nonmagnetic spacer layer between the reference layer and the free layer. The free layer has stable magnetic states and is programmable using spin-transfer torque and/or spin-orbit interaction torque. The first magnetic junction of a pair receives the input signal. The second magnetic junction of the pair receives the input signal complement. The output transistors are coupled with the magnetic junctions such that each pair of magnetic junctions forms a voltage divider. The output transistors form a sense amplifier.

Method and system for providing a variation resistant magnetic junction-based XNOR cell usable in neuromorphic computing

A field effect transistor (FET) for an nFET and/or a pFET device including a substrate and a fin including at least one channel region decoupled from the substrate. The FET also includes a source electrode and a drain electrode on opposite sides of the fin, and a gate stack extending along a pair of sidewalls of the channel region of the fin. The gate stack includes a gate dielectric layer and a metal layer on the gate dielectric layer. The FET also includes an oxide separation region separating the channel region of the fin from the substrate. The oxide separation region includes a dielectric material that includes a portion of the gate dielectric layer of the gate stack. The oxide separation region extends completely from a surface of the channel region facing the substrate to a surface of the substrate facing the channel region.

Field effect transistor with decoupled channel and methods of manufacturing the same

A method of manufacturing a field effect transistor includes forming a fin on a substrate, forming source and drain electrodes on opposite sides of the fin, forming a gate stack on a channel portion of the fin between the source and drain electrodes, forming gate spacers on extension portions of the fin on opposite sides of the gate stack, removing at least portions of the gate spacers to expose the extension portions of the fin, and hydrogen annealing the extension portions of the fin. Following the hydrogen annealing of the extension portions of the fin, the channel portion of the fin has a first width and the extension portions of the fin have a second width greater than the first width.

FinFET with reduced extension resistance and methods of manufacturing the same

A method of manufacturing a field effect transistor includes forming a fin on a substrate, forming source and drain electrodes on opposite sides of the fin, forming a gate stack on a channel portion of the fin between the source and drain electrodes, forming gate spacers on extension portions of the fin on opposite sides of the gate stack, removing at least a portion of the gate spacers to expose the extension portions of the fin, and thinning the extension portions of the fin. Following the thinning of the extension portions of the fin, the channel portion of the fin has a first width and the extension portions of the fin have a second width less than the first width.

Growing community of inventors

Leander, TX, United States of America

Borna Josip Obradovic