KR102561799B1 - 디바이스에서 딥러닝 모델의 레이턴시를 예측하는 방법 및 시스템 - Google Patents

디바이스에서 딥러닝 모델의 레이턴시를 예측하는 방법 및 시스템 Download PDF

Info

Publication number
KR102561799B1
KR102561799B1 KR1020220094417A KR20220094417A KR102561799B1 KR 102561799 B1 KR102561799 B1 KR 102561799B1 KR 1020220094417 A KR1020220094417 A KR 1020220094417A KR 20220094417 A KR20220094417 A KR 20220094417A KR 102561799 B1 KR102561799 B1 KR 102561799B1
Authority
KR
South Korea
Prior art keywords
latency
neural network
network layer
single neural
deep learning
Prior art date
Application number
KR1020220094417A
Other languages
English (en)
Korean (ko)
Other versions
KR20230024835A (ko
Inventor
김정호
김민수
김태호
Original Assignee
주식회사 노타
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 주식회사 노타 filed Critical 주식회사 노타
Priority to US17/819,281 priority Critical patent/US12033052B2/en
Publication of KR20230024835A publication Critical patent/KR20230024835A/ko
Application granted granted Critical
Publication of KR102561799B1 publication Critical patent/KR102561799B1/ko

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • G06N20/20—Ensemble learning
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/044—Recurrent networks, e.g. Hopfield networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Software Systems (AREA)
  • Computing Systems (AREA)
  • Artificial Intelligence (AREA)
  • Mathematical Physics (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • Biomedical Technology (AREA)
  • Molecular Biology (AREA)
  • General Health & Medical Sciences (AREA)
  • Computational Linguistics (AREA)
  • Biophysics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Medical Informatics (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
KR1020220094417A 2021-08-12 2022-07-29 디바이스에서 딥러닝 모델의 레이턴시를 예측하는 방법 및 시스템 KR102561799B1 (ko)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US17/819,281 US12033052B2 (en) 2021-08-12 2022-08-11 Latency prediction method and computing device for the same

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
KR1020210106527 2021-08-12
KR20210106527 2021-08-12

Publications (2)

Publication Number Publication Date
KR20230024835A KR20230024835A (ko) 2023-02-21
KR102561799B1 true KR102561799B1 (ko) 2023-07-31

Family

ID=85200773

Family Applications (1)

Application Number Title Priority Date Filing Date
KR1020220094417A KR102561799B1 (ko) 2021-08-12 2022-07-29 디바이스에서 딥러닝 모델의 레이턴시를 예측하는 방법 및 시스템

Country Status (2)

Country Link
KR (1) KR102561799B1 (fr)
WO (1) WO2023017884A1 (fr)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116668475B (zh) * 2023-05-18 2023-12-26 尚学仕教育科技(北京)有限公司 在线教育操作系统

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20210073242A (ko) * 2019-12-10 2021-06-18 삼성전자주식회사 모델 최적화 방법 및 장치 및 모델 최적화 장치를 포함한 가속기 시스템

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
Matthias Wess et al., "ANNETTE: ACCURATE NEURAL NETWORK EXECUTION TIME ESTIMATION WITH STACKED MODELS," arXiv:2105.03176v1 [cs.LG] 7 May 2021 (2021.05.07.)*

Also Published As

Publication number Publication date
KR20230024835A (ko) 2023-02-21
WO2023017884A1 (fr) 2023-02-16

Similar Documents

Publication Publication Date Title
CN111406267B (zh) 使用性能预测神经网络的神经架构搜索
EP3446260B1 (fr) Rétropropagation dans le temps, économe en mémoire
CN110651280B (zh) 投影神经网络
CN110807515B (zh) 模型生成方法和装置
CN110852438B (zh) 模型生成方法和装置
US11861474B2 (en) Dynamic placement of computation sub-graphs
CN110476172A (zh) 用于卷积神经网络的神经架构搜索
CN110766142A (zh) 模型生成方法和装置
US11386256B2 (en) Systems and methods for determining a configuration for a microarchitecture
CN110366734A (zh) 优化神经网络架构
US20220230048A1 (en) Neural Architecture Scaling For Hardware Accelerators
CN111340221B (zh) 神经网络结构的采样方法和装置
US20210089834A1 (en) Imagination-based agent neural networks
CN114330699A (zh) 神经网络结构搜索方法及装置
CN111340220B (zh) 用于训练预测模型的方法和装置
WO2021224720A1 (fr) Détermination de dépendances de données en série chronologique à variables multiples
CN109858615A (zh) 具有记忆的低通递归神经网络系统
CN111353601A (zh) 用于预测模型结构的延时的方法和装置
CN114072809A (zh) 经由神经架构搜索的小且快速的视频处理网络
KR102561799B1 (ko) 디바이스에서 딥러닝 모델의 레이턴시를 예측하는 방법 및 시스템
KR20220094564A (ko) 딥러닝 모델 서빙 최적화를 위한 모델 자동 경량화 방법 및 장치, 이를 이용한 클라우드 추론 서비스 제공 방법
CN116897356A (zh) 算子的调度运行时间比较方法、装置及存储介质
KR20230024950A (ko) 최적 파라미터 결정 방법 및 시스템
CN110782016A (zh) 用于优化神经网络架构搜索的方法和装置
US20220405649A1 (en) Quantum machine learning model feature space generation

Legal Events

Date Code Title Description
GRNT Written decision to grant