Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[12] Autoformer: Decomposition transformers with auto-correlation for long-term series forecasting #12

Open
techzzt opened this issue Sep 20, 2022 · 0 comments

Comments

@techzzt
Copy link
Owner

techzzt commented Sep 20, 2022

Autoformer: Decomposition transformers with auto-correlation for long-term series forecasting

image

  • Auto-correlation과 time-delay aggregation을 활용한 모델 제안
  • Correlation을 고려하면서 time series 데이터 간의 dependency를 반영하고 information aggregation을 통해 정보를 추출
  • Decoder에서는 seasonal과 trend로 데이터를 구분하고 이를 반영

image

  • Fast Fourier Transform을 통해 연산을 효율적으로 수행
  • 또한, time delay aggregation 과정에서 rolling된 데이터의 auto correlation을 계산하기 위해 FFT 사용
  • Aucto correlation 계산 과정을 통해 time delay aggregation으로부터의 sub series 유사도와 autocorrelation을 결합한 period-based dependency를 계산

image

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant