# US 2020/0394502 A1 · Neuromorphic computing device

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2020/0394502 A1
- Document ID: US20200394502A1
- Status: Published application
- Kind: A1
- Application number: 16/441,106
- Filed: 2019-06-14
- Published: 2020-12-17
- Earliest claimed: 2019-06-14
- Assignees: Macronix International Co., Ltd.
- Applicants: Macronix International Co., Ltd.
- Inventors: Shih-Hung CHEN
- Web page: https://patentgrep.com/patent/US20200394502A1

## CPC

- G06N 3/063: using electronic means
- G06N 3/065: Analogue means
- G06N 3/045: Combinations of networks
- G06N 3/0499: Feedforward networks
- G06N 3/082: modifying the architecture, e.g. adding, deleting or silencing nodes or connections
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

A neuromorphic computing device includes first neural circuits, second neural circuits and synapse weights. The first neural circuits are disposed in a first neural region. The second neural circuits are disposed in a second neural region. The synapse weights are electrically connected between the first neural circuits and the second neural circuits, and disposed in a synapse region. The first neural region and the second neural region are on opposing sides of the synapse region respectively.

## Claims

### Claim 1

A neuromorphic computing device, comprising:
first neural circuits disposed in a first neural region;
second neural circuits disposed in a second neural region; and
synapse weights electrically connected between the first neural its and the second neural circuits, and disposed in a synapse region, wherein the first neural region and the second neural region are on opposing sides of the synapse region respectively.

### Claim 2 (depends on 1)

The neuromorphic computing device according to claim 1, comprising resistive material layers and insulating layers stacked alternately, wherein one of the synapse weights comprises at least one of the resistive material layers.

### Claim 3 (depends on 1)

The neuromorphic computing device according to claim 1, comprising transistors on the same plane, wherein one of the synapse weights comprises at least one of transistors.

### Claim 4 (depends on 1)

The neuromorphic computing device according to claim 1, wherein each of the synapse weights comprises a signal input terminal and a signal output terminal, the signal input terminals are between the first neural circuits and the signal output terminals, the signal output terminals are between the second neural circuits and the signal input terminals.

### Claim 5 (depends on 1)

The neuromorphic computing device according to claim 1, wherein the synapse weights are defined into different synapse weight groups electrically connected to one of the first neural circuits, and one of the second neural circuits respectively.

### Claim 6 (depends on 1)

The neuromorphic computing device according to claim 1, wherein one of the first neural circuits arranged along a direction is electrically connected to a first group of synapse weights arranged along an another direction of the synapse weights, a second group of synapse weights arranged along the direction of the synapse weights is electrically connected to one of the second neural circuits arranged along the direction.

### Claim 7 (depends on 6)

The neuromorphic computing device according to claim 6, wherein each of the synapse weights comprises a resistor, extended along a yet another direction.

### Claim 8 (depends on 1)

The neuromorphic computing device according to claim 1, wherein one of the first neural circuits arranged along a direction is electrically connected to a first group of synapse weights arranged along the direction of the synapse weights, a second group of synapse weights arranged along an another direction of the synapse weights is electrically connected to one of the second neural circuits arranged along the direction.

### Claim 9 (depends on 7)

The neuromorphic computing device according to claim 7, wherein each of the synapse weights comprises a resistor, extended along a yet another direction.

### Claim 10 (depends on 1)

The neuromorphic computing device according to claim 1, wherein the synapse region, the first neural region and the second neural region are non-overlapping with each other.

### Claim 11 (depends on 1)

The neuromorphic computing device according to claim 1, further comprising a substrate under the first neural circuits, the second neural circuits, and the synapse weights.

### Claim 12

A neuromorphic computing device, comprising:
a substrate;
synapse weights on the substrate; and
neural circuits electrically connected to the synapse weights, and disposed on a side of the synapse weights facing towards the substrate.

### Claim 13 (depends on 12)

The neuromorphic computing device according to claim 12, wherein the neural circuits comprise first neural circuits disposed at the side of the synapse weights facing towards the substrate.

### Claim 14 (depends on 13)

The neuromorphic computing device according to claim 13, wherein the neural circuits further comprise second neural circuits disposed at the side of the synapse weights facing towards the substrate, the synapse weights are electrically connected between the first neural circuits and the second neural circuits.

### Claim 15 (depends on 12)

The neuromorphic computing device according to claim 12, wherein the neural circuits comprises second neural circuits disposed at the side of the synapse weights facing towards the substrate.

### Claim 16 (depends on 12)

The neuromorphic computing device according to claim 12, wherein each of the synapse weights comprises a signal input terminal and a signal output terminal, one of the signal input terminal and the signal output terminal is at the side of the synapse weights facing towards the substrate, the other of the signal input terminal and the signal output terminal is on an another side of the synapse weights back to the substrate.

### Claim 17 (depends on 12)

The neuromorphic computing device according to claim 12, wherein the neural circuits comprise first neural circuits and second neural circuits, each of the synapse weights comprises a signal input terminal and a signal output terminal, neural signals from the first neural circuits are transferred into the synapse weights through the signal input terminals and converted into weighted signals by the synapse weights, and then the weighted signals are transferred from the synapse weights through the signal output terminal to the second neural circuits.

### Claim 18 (depends on 12)

The neuromorphic computing device according to claim 12, wherein the synapse weights are defined into different synapse weight groups electrically connected to one of the first neural circuits, and one of the second neural circuits respectively.

### Claim 19 (depends on 12)

The neuromorphic computing device according to claim 12, comprising resistive material layers and insulating layers stacked alternately, wherein one of the synapse weights comprises at least one of the resistive material layers.

### Claim 20 (depends on 12)

The neuromorphic computing device according to claim 12, comprising transistors on the same plane, wherein one of the synapse weights comprises at least one of transistors
