JPWO2020230374A5 - - Google Patents
Download PDFInfo
- Publication number
- JPWO2020230374A5 JPWO2020230374A5 JP2021519259A JP2021519259A JPWO2020230374A5 JP WO2020230374 A5 JPWO2020230374 A5 JP WO2020230374A5 JP 2021519259 A JP2021519259 A JP 2021519259A JP 2021519259 A JP2021519259 A JP 2021519259A JP WO2020230374 A5 JPWO2020230374 A5 JP WO2020230374A5
- Authority
- JP
- Japan
- Prior art keywords
- product
- unit
- sum
- output
- arithmetic unit
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 230000001186 cumulative effect Effects 0.000 claims 18
- 238000004364 calculation method Methods 0.000 claims 8
- 238000013528 artificial neural network Methods 0.000 claims 1
- 238000005259 measurement Methods 0.000 claims 1
Applications Claiming Priority (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2019089422 | 2019-05-10 | ||
| JP2019089422 | 2019-05-10 | ||
| PCT/JP2020/003485 WO2020230374A1 (ja) | 2019-05-10 | 2020-01-30 | 演算装置および演算システム |
Publications (3)
| Publication Number | Publication Date |
|---|---|
| JPWO2020230374A1 JPWO2020230374A1 (https=) | 2020-11-19 |
| JPWO2020230374A5 true JPWO2020230374A5 (https=) | 2022-02-09 |
| JP7435602B2 JP7435602B2 (ja) | 2024-02-21 |
Family
ID=73289562
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| JP2021519259A Active JP7435602B2 (ja) | 2019-05-10 | 2020-01-30 | 演算装置および演算システム |
Country Status (5)
| Country | Link |
|---|---|
| US (1) | US20220300253A1 (https=) |
| EP (1) | EP3968242A4 (https=) |
| JP (1) | JP7435602B2 (https=) |
| CN (1) | CN113811900A (https=) |
| WO (1) | WO2020230374A1 (https=) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20210334072A1 (en) * | 2020-04-22 | 2021-10-28 | Facebook, Inc. | Mapping convolution to connected processing elements using distributed pipelined separable convolution operations |
| KR102891573B1 (ko) * | 2020-06-30 | 2025-11-26 | 삼성전자주식회사 | 텐서 처리 방법, 가속기 및 이를 포함한 가속기 시스템 |
| JP7659373B2 (ja) * | 2020-07-09 | 2025-04-09 | キヤノン株式会社 | 畳み込みニューラルネットワーク処理装置 |
| US12056459B2 (en) * | 2021-07-02 | 2024-08-06 | Qualcomm Incorporated | Compute in memory architecture and dataflows for depth-wise separable convolution |
| FR3159250A1 (fr) * | 2024-02-09 | 2025-08-15 | Neurxcore | Circuit électronique de type multiplieur-accumulateur (MAC) |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP6700712B2 (ja) * | 2015-10-21 | 2020-05-27 | キヤノン株式会社 | 畳み込み演算装置 |
| US10360470B2 (en) | 2016-10-10 | 2019-07-23 | Gyrfalcon Technology Inc. | Implementation of MobileNet in a CNN based digital integrated circuit |
| US10083171B1 (en) * | 2017-08-03 | 2018-09-25 | Gyrfalcon Technology Inc. | Natural language processing using a CNN based integrated circuit |
| JP2019040403A (ja) * | 2017-08-25 | 2019-03-14 | ルネサスエレクトロニクス株式会社 | 半導体装置および画像認識システム |
| CN107844828B (zh) * | 2017-12-18 | 2021-07-30 | 南京地平线机器人技术有限公司 | 神经网络中的卷积计算方法和电子设备 |
| CN108108811B (zh) * | 2017-12-18 | 2021-07-30 | 南京地平线机器人技术有限公司 | 神经网络中的卷积计算方法和电子设备 |
| CN109284817B (zh) * | 2018-08-31 | 2022-07-05 | 中国科学院上海高等研究院 | 深度可分离卷积神经网络处理架构/方法/系统及介质 |
-
2020
- 2020-01-30 EP EP20806680.3A patent/EP3968242A4/en not_active Withdrawn
- 2020-01-30 JP JP2021519259A patent/JP7435602B2/ja active Active
- 2020-01-30 US US17/607,953 patent/US20220300253A1/en not_active Abandoned
- 2020-01-30 CN CN202080032634.9A patent/CN113811900A/zh not_active Withdrawn
- 2020-01-30 WO PCT/JP2020/003485 patent/WO2020230374A1/ja not_active Ceased
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JPWO2020230374A5 (https=) | ||
| CN110383300B (zh) | 一种计算装置及方法 | |
| EP3373210B1 (en) | Transposing neural network matrices in hardware | |
| JP5408913B2 (ja) | 高速かつ効率的な行列乗算ハードウェアモジュール | |
| JP2022533221A (ja) | 行列演算用の装置及び方法 | |
| EP4095719A1 (en) | Sparse matrix multiplication in hardware | |
| Kahar et al. | High speed vedic multiplier used vedic mathematics | |
| KR20010020446A (ko) | 신경 프로세서, 포화 유니트, 계산 유니트 및 가산기 회로 | |
| CN107797962A (zh) | 基于神经网络的计算阵列 | |
| WO2003056513A1 (en) | Integrated circuit for universal computing with near zero programming complexity | |
| TWI808259B (zh) | 具有壓縮進位之數位電路 | |
| JPS6347874A (ja) | 算術演算装置 | |
| Nedjah et al. | Compact yet efficient hardware implementation of artificial neural networks with customized topology | |
| JP7700142B2 (ja) | 機械学習アクセラレータの電力削減 | |
| JP2012528391A5 (https=) | ||
| US10831445B1 (en) | Multimodal digital multiplication circuits and methods | |
| CN120569716A (zh) | 用于矩阵乘法的多模态脉动阵列 | |
| CN1020170C (zh) | 高速数字处理器 | |
| KR20220031098A (ko) | 부호형 다중-워드 곱셈기 | |
| CN111930674B (zh) | 乘累加运算装置及方法、异构智能处理器及电子设备 | |
| Basiri et al. | Multiplication acceleration through quarter precision Wallace tree multiplier | |
| CN118318237A (zh) | 乘积累加单元中的模空间处理 | |
| JPS6373472A (ja) | 積和演算方法 | |
| Ago et al. | The parallel FDFM processor core approach for neural networks | |
| KR102848552B1 (ko) | 부동소수점 연산기를 이용한 엠에이씨 장치 및 그 제어 방법 |