Abstract: Described is an ultra-dense ferroelectric memory. The memory is fabricated using a patterning method by that applies atomic layer deposition with selective dry and/or wet etch to increase memory density at a given via opening. A ferroelectric capacitor in one example comprises: a first structure (e.g., first electrode) comprising metal; a second structure (e.g., a second electrode) comprising metal; and a third structure comprising ferroelectric material, wherein the third structure is between and adjacent to the first and second structures, wherein a portion of the third structure is interdigitated with the first and second structures to increase surface area of the third structure. The increased surface area allows for higher memory density.

Abstract: Disclosed herein are capacitors with ferroelectric or antiferroelectric (FE/AFE) material and dielectric material, as well as related methods and devices. In some embodiments, a capacitor may include two electrodes, a layer of FE/AFE material between the electrodes, and a layer of dielectric material between the electrodes.

Abstract: Techniques are disclosed for forming vias for integrated circuit structures. During an additive via formation process, a dielectric material is deposited, an etch stop layer is deposited, a checkerboard pattern is deposited on the etch stop layer, regions in the checkerboard pattern are removed where it is desired to have vias, openings are etched in the dielectric material through the removed regions, and the openings are filled with a first via material. This is then repeated for a second via material. During the subtractive via formation process, a first via material is deposited, an etch stop layer is deposited, a checkerboard pattern is deposited on the etch stop layer, regions in the checkerboard pattern are removed where it is not desired to have vias, openings are etched in the first via material through the removed regions. This is then repeated for a second via material.

Abstract: Some embodiments include apparatuses having a first magnet, a first stack of layers coupled to a first portion of the first magnet, a first layer coupled to a second portion of the first magnet, a second magnet, a second stack of layers coupled to a first portion of the second magnet, a second layer coupled to a second portion of the second magnet, a conductor coupled to the first stack of layers and to the second layer, and a conductive path coupled to the first portion of the first magnet and to the first portion of the second magnet, each of the first and second layers including a magnetoelectric material, each of the first and second stacks of layers providing an inverse spin orbit coupling effect.

Abstract: A memory circuit has compute-in-memory circuitry that enables a multiply-accumulate (MAC) operation based on shared charge. Row access circuitry drives multiple rows of a memory array to multiply a first data word with a second data word stored in the memory array. The row access circuitry drives the multiple rows based on the bit pattern of the first data word. Column access circuitry drives a column of the memory array when the rows are driven. Accessed rows discharge the column line in an accumulative fashion. Sensing circuitry can sense voltage on the column line. A processor in the memory circuit computes a MAC value based on the voltage sensed on the column.

Abstract: One embodiment provides an apparatus. The apparatus includes a pair of nonvolatile resistive random access memory (RRAM) memory cells coupled to a volatile static RAM (SRAM) memory cell. The pair of nonvolatile RRAM memory cells includes a first RRAM memory cell and a second RRAM memory cell. The first RRAM memory cell includes a first resistive memory element coupled to a first bit line, and a first selector transistor coupled between the first resistive memory element and a first output node of the volatile SRAM memory cell. The second RRAM memory cell includes a second resistive memory element coupled to a second bit line, and a second selector transistor coupled between the second resistive memory element and a second output node of the volatile SRAM memory cell.

Abstract: Techniques are provided for implementing capsule neural networks (NNs) using vector spin neurons. A vector spin neuron according to an embodiment includes a first magnet, polarized in a first direction, to receive a first input current. The first input current is based on an NN input value and weighting factor. The vector spin neuron also includes a second magnet, polarized in a direction orthogonal to the first direction, to receive a second input current. The second input current is based on a second NN input value and weighting factor. The first and second magnets generate spin polarized currents. In some such embodiments, the vector spin neuron further includes a third magnet, which is unpolarized, and a conductor to couple output regions of the first and second magnets to an input region of the third magnet. The third magnet applies a non-linear activation function to the sum of the spin polarized currents.

Abstract: One embodiment provides an apparatus. The apparatus includes a first transistor and a second transistor. The first transistor includes a first drain, a first source coupled to the first drain by a first channel, and a first gate stack comprising a plurality of layers. The second transistor includes a second drain, a second source coupled to the second drain by a second channel, and a second gate stack comprising a plurality of layers. Each gate stack includes a work function material layer to optimize a threshold voltage variation between the transistors.

Abstract: Techniques are provided for implementing a hybrid processing architecture comprising a general-purpose processor (CPU) coupled to an analog in-memory artificial intelligence (AI) processor. A hybrid processor implementing the techniques according to an embodiment includes an AI processor configured to perform analog in-memory computations based on neural network (NN) weighting factors and input data provided by the CPU. The AI processor includes one or more NN layers. The NN layers include digital access circuits to receive data and weighting factors and to provide computational results. The NN layers also include memory circuits to store data and weights, and further include bit line processors and cross bit line processors to perform analog dot product computations between columns of the data memory circuits and the weight factor memory circuits. Some of the NN layers are configured as convolutional NN layers and others are configured as fully connected NN layers, according to some embodiments.

Abstract: Techniques are provided for implementing a hybrid processing architecture comprising a general-purpose processor (CPU) and a neural processing unit (NPU), coupled to an analog in-memory artificial intelligence (AI) processor. According to an embodiment, the hybrid processor implements an AI instruction set including instructions to perform analog in-memory computations. The AI processor comprises one or more layers, the NN layers including memory circuitry and analog processing circuitry. The memory circuitry is configured to store the weighting factors and the input data. The analog processing circuitry is configured to perform analog calculations on the stored weighting factors and the stored input data in accordance with the execution, by the NPU, of instruction from the AI instruction set. The AI instruction set includes instructions to perform dot products, multiplication, differencing, normalization, pooling, thresholding, transposition, and backpropagation training.

Abstract: Described is an apparatus which comprises: a word line; a source line; a bit-line; and a memory bit-cell coupled to the source line, the bit-line, and the word line, wherein the memory bit-cell comprises a capacitor including ferroelectric material and a transistor fabricated on a backend of a die.

Abstract: Described is an apparatus which comprises: a first layer comprising a semiconductor; a second layer comprising an material, the second layer adjacent to the first layer; a third layer comprising a high-k insulating material, the third layer adjacent to the second layer; a fourth layer comprising a ferroelectric material, the fourth layer adjacent to the third layer; and a fifth layer comprising a high-k insulating material, the fifth layer adjacent to the fourth layer.

Abstract: A binary CIM circuit enables all memory cells in a memory array to be effectively accessible simultaneously for computation using fixed pulse widths on the wordlines and equal capacitance on the bitlines. The fixed pulse widths and equal capacitance ensure that a minimum voltage drop in the bitline represents one least significant bit (LSB) so that the bitline voltage swing remains safely within the maximum allowable range. The binary CIM circuit maximizes the effective memory bandwidth of a memory array for a given maximum voltage range of bitline voltage.

Abstract: Described is an apparatus which comprises a transistor including: a layer of ferroelectric material; a layer of insulating material; and an oxide layer or a metal layer sandwiched between the layer of ferroelectric material and the layer of insulating material, wherein thickness of the ferroelectric material is less than thickness of the layer of insulating material; and a driver coupled to the transistor. Described is an apparatus which comprises: a transistor including: a first oxide layer of High-K material; a second oxide layer; and a layer of nanocrystals sandwiched between the first and second oxide layers, wherein thickness of first oxide layer is greater than thickness of the second oxide layer; and a driver coupled to the transistor.

Abstract: An embodiment includes a system comprising: first, second, and third word lines on a semiconductor material; first, second, and third channels; first, second, and third capacitors including a ferroelectric material; a bit line; first, second, third, fourth, and fifth semiconductor nodes, wherein the first semiconductor node couples the first capacitor to the first channel, the second semiconductor node couples the bit line to the first channel; the third semiconductor node couples the second capacitor to the second channel, the fourth semiconductor node couples the third capacitor to the third channel, and the fifth semiconductor node couples the bit line to the third channel; wherein the first channel has a long axis and a short axis; wherein the long axis intersects a continuous, uninterrupted portion of the semiconductor material from the first channel to the third channel.

Abstract: A memory device includes a spin orbit electrode structure having a dielectric structure including a first sidewall, a second sidewall opposite to the first sidewall, a top surface. The spin orbit electrode structure further includes an electrode having a spin orbit material adjacent to the dielectric structure, where the electrode has a first electrode portion on the top surface, a second electrode portion adjacent to the first sidewall and a third electrode portion adjacent to the second sidewall. The first electrode portion, the second electrode portion and the third electrode portion are contiguous. The spin orbit electrode structure further includes a conductive interconnect in contact with the second electrode portion or the third electrode portion. The memory device further includes a magnetic junction device on a portion of the top surface of the first electrode portion.

Abstract: A perpendicular spin orbit memory device includes a first electrode having a magnetic material and platinum and a material layer stack on a portion of the first electrode. The material layer stack includes a free magnet, a fixed magnet above the first electrode, a tunnel barrier between the free magnet and the fixed magnet and a second electrode coupled with the fixed magnet.

Abstract: A perpendicular spin orbit torque memory device includes a first electrode having tungsten and at least one of nitrogen or oxygen and a material layer stack on a portion of the first electrode. The material layer stack includes a free magnet, a fixed magnet above the first magnet, a tunnel barrier between the free magnet and the fixed magnet and a second electrode coupled with the fixed magnet.

Abstract: The present disclosure is directed to systems and methods of implementing a neural network using in-memory mathematical operations performed by pipelined SRAM architecture (PISA) circuitry disposed in on-chip processor memory circuitry. A high-level compiler may be provided to compile data representative of a multi-layer neural network model and one or more neural network data inputs from a first high-level programming language to an intermediate domain-specific language (DSL). A low-level compiler may be provided to compile the representative data from the intermediate DSL to multiple instruction sets in accordance with an instruction set architecture (ISA), such that each of the multiple instruction sets corresponds to a single respective layer of the multi-layer neural network model. Each of the multiple instruction sets may be assigned to a respective SRAM array of the PISA circuitry for in-memory execution.

Abstract: Described is an apparatus which comprises: a first p-type Tunneling Field-Effect Transistor (TFET); a first n-type TFET coupled in series with the first p-type TFET; a first node coupled to gate terminals of the first p-type and n-type TFETs; a first clock node coupled to a source terminal of the first TFET, the first clock node is to provide a first clock; and a second clock node coupled to a source terminal of the second TFET, the second clock node is to provide a second clock.