(A) hybrid FP-FXP precision deep learning processor with outlier compensation for image-to-image application이미지 변환을 위한 이상치 보정을 적용한 하이브리드 부동-고정점 소수점 딥러닝 프로세서

Cited 0 time in webofscience Cited 0 time in scopus
  • Hit : 86
  • Download : 0
DC FieldValueLanguage
dc.contributor.advisorYoo, Hoi-Jun-
dc.contributor.advisor유회준-
dc.contributor.authorLi, Zhiyong-
dc.date.accessioned2022-04-27T19:31:26Z-
dc.date.available2022-04-27T19:31:26Z-
dc.date.issued2021-
dc.identifier.urihttp://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=963442&flag=dissertationen_US
dc.identifier.urihttp://hdl.handle.net/10203/296023-
dc.description학위논문(석사) - 한국과학기술원 : 전기및전자공학부, 2021.8,[iii, 19 p. :]-
dc.description.abstractA Hybrid floating-point (FP) and fixed-point (FXP) deep learning processor with an outlier-aware channel splitting algorithm is proposed for image-to-image applications on mobile devices. Since the high quality of the reconstructed image through deep learning based image-to-image application requires high bit-precision (> FP16), the mobile processor suffers from the high computation power and large external memory access (EMA). In this work, the proposed algorithm reduces 16-bit FP data to 8-bit FXP data, and only few outliers (< 10%) are computed in 16-bit FP while maintaining the image reconstruction quality. Therefore, it reduces EMA by 45.5%. Moreover, the hierarchical processor accelerates these dense 8-bit FXP data and sparse 16-bit FP data, and the functional L2 memory aggregates the convolution output of them by forming the pipeline, which reduces 98% of latency. The proposed system is simulated in 28nm COMS technology, and it occupies 4.16mm2. The hierarchical processor successfully demonstrates the × 4 scale Full-HD super-resolution generation achieving 76 frames-per-second (fps) with 133.3 mW power-consumption at 0.9 V supply and 3.6 TOPS/W of energy-efficiency which is × 3.27 higher than the previous 16-bit FXP processor.-
dc.languageeng-
dc.publisher한국과학기술원-
dc.subjectConvolutional neural network (CNN)▼aoutlier-aware▼achannel split▼ahybrid precision▼aimage-to-image▼asuper-resolution▼astyle transfer▼anon-sparse CNN-
dc.subject합성 곱 네트워크▼a이상치 보정▼a채널 분할▼a하이브리드 정밀도▼a이미지-이미지간 변환▼a초 해상도 알고리즘▼a스타일 변환▼a비 희소성 합성 곱 신경망-
dc.title(A) hybrid FP-FXP precision deep learning processor with outlier compensation for image-to-image application-
dc.title.alternative이미지 변환을 위한 이상치 보정을 적용한 하이브리드 부동-고정점 소수점 딥러닝 프로세서-
dc.typeThesis(Master)-
dc.identifier.CNRN325007-
dc.description.department한국과학기술원 :전기및전자공학부,-
dc.contributor.alternativeauthor이지용-
Appears in Collection
EE-Theses_Master(석사논문)
Files in This Item
There are no files associated with this item.

qr_code

  • mendeley

    citeulike


rss_1.0 rss_2.0 atom_1.0