| 발표자 | 조영성 |
|---|---|
| 발표일자 | 2022-02-22 |
| 저자 | Yanping Huang, Youlong Cheng, Ankur Bapna, Orhan Firat, Mia Xu Chen, Dehao Chen, HyoukJoong Lee, Jiquan Ngiam, Quoc V. Le, Yonghui Wu, Zhifeng Chen |
| 학회명 | |
| 논문지 | https://arxiv.org/abs/1811.06965 |
댓글 2
-
이지형
2022.02.22 20:20
우리가 관심을 갖지 않았던 분야를 소개하였음. 우리 연구에도 필요한 개념으로 많이 활용할 수 있도록 잘 알아 둘 필요가 있음 -
윤수현
2022.02.22 20:27
Multi-GPU 환경에서의 Pipeline parallelism에 대한 소개였습니다. 실험 코드를 작성할 때에도 parallelism을 고려해야겠다는 필요성을 느끼게 되네요. 좋은 발표 감사합니다. Distributed Data parallelism 환경에서는 Batchnorm 대신 SyncBN을 사용하는데, GPipe와 같은 pipeline parallelism을 사용할 때도 학습에 변화를 주어야 하는 부분이 있을까요?
2022
Rep VGG: Making VGG-style ConvNet Again
2022.07.14
발표자: 장성인
발표일자: 2022-07-13
저자: Xiaohan Ding
학회명: CVPR
Perturbed and Strict Mean Teachers for Semi-supervised Semantic Segmentation
2022.07.13
발표자: 윤수현
발표일자: 2022-07-13
저자: Yuyuan Liu, Yu Tian, Yuanhong Chen, Fengbei Liu, Vasileios Belagiannis, Gustavo Carneiro
학회명: CVPR 2022
CodeT5: Identifier-aware Unified Pre-trained Encoder-Decoder Models for Code Understanding and Generation
2022.07.13
발표자: 나철원
발표일자: 2022-07-13
저자: Yue Wang, Weishi Wang, Shafiq Joty, and Steven C.H. Hoi
학회명: EMNLP 2021
On the Evaluation of Neural Code Summarization
2022.06.29
발표자: 최윤석
발표일자: 2022-06-29
저자: Ensheng Shi; Yanlin Wang; Lun Du; Junjie Chen; Shi Han; Hongyu Zhang; Dongmei Zhang; Hongbin Sun
학회명: ICSE 2022
Consistent Explanations by Contrastive Learning
2022.06.29
발표자: 김누리
발표일자: 2022-06-29
저자: Pillai, Vipin, et al.
학회명: CVPR 2022
Large Loss Matters in Weakly Supervised Multi-Label Classification
2022.06.29
발표자: 송경렬
발표일자: 2022-06-29
저자: Youngwook Kim1* , Jae Myung Kim2*, Zeynep Akata2,3,4, Jungwoo Lee1,5
학회명: CVPR 2022
Improved Transformer for High-Resolution GANs
2022.06.22
발표자: 신재민
발표일자: 2022-06-22
저자: Long Zhao
학회명: NeurIPS 2021
Memorizing Normality to Detect Anomaly: Memory-augmented Deep Autoencoder for Unsupervised Anomaly Detection
2022.06.22
발표자: 지혜림
발표일자: 2022-06-22
저자: Dong Gong1 , Lingqiao Liu1 , Vuong Le2 , Budhaditya Saha2 , Moussa Reda Mansour3 , Svetha Venkatesh2 , Anton van den Hengel
학회명: ICCV 2019
Noisy Boundaries: Lemon or Lemonade for Semi-supervised Instance Segmentation?
2022.06.22
발표자: 임지영
발표일자: 2022-06-22
학회명: CVPR 2022
Contrast Learning Visual Attention for Multi Label Classification
2022.06.15
발표자: 이진섭
발표일자: 2022-06-15
저자: Son D.Dao, Ethan Zhao, Dinh Phung, Jianfei Cai
학회명: Arxiv.
C2L: Causally Contrastive Learning for Robust Text Classification
2022.06.15
발표자: 김효준
발표일자: 2022-06-15
저자: Seungtaek Choi, Myeongho Jeong, Hojae Han, Seung won Hwang
학회명: AAAI 2022
Effectively Leveraging Attributes for Visual Similarity
2022.06.14
발표자: 박은미
발표일자: 2022-06-14
저자: Samarth Mishra, Zhongping Zhang, Yuan Shen, Ranjitha Kumar, Venkatesh Saligrama, Bryan Plummer
학회명: ICCV2021
CORE: Simple and Effective Session-based Recommendation within Consistent Representation Space
2022.06.14
발표자: 양희윤
발표일자: 2022-06-14
저자: Yupeng Hou, Binbin Hu, Zhipiang Zhang, Wayne Xin Zhao
학회명: SIGIR 2022
FedPara: Low-rank Hadamard Product for Communication-Efficient Federated Learning
2022.06.09
발표자: 강용훈
발표일자: 2022-06-10
저자: Nam Hyeon-Woo, Moon Ye-Bin, Tae-Hyun Oh
학회명: ICLR 2022
Do Vision Transformers See Like Convolutional Neural Networks?
2022.06.10
발표자: 안재한
발표일자: 2022-06-10
저자: Maithra Raghu, Thomas Unterthiner, Simon Kornblith, Chiyuan Zhang, Alexey Dosovitskiy
학회명: NeurIPS2021
Large Language Models are Zero-Shot Reasoners
2022.06.10
발표자: 홍만수
발표일자: 2022-06-10
저자: Takeshi Kojima, Shixiang Shane Gu, Machel Reid, Yutaka Matsuo, Yusuke Iwasawa
When Chosen Wisely, More Data Is What You Need: A Universal Sample-Efficient Strategy For Data Augmentation
2022.05.09
발표자: 김호승
발표일자: 2022-05-09
학회명: ACL2022
Likelihood Ratios for Out-of-Distribution Detection
2022.05.09
발표자: 노순철
발표일자: 2022-05-09
저자: Jie Ren, et al.
학회명: NeurIPS 2019
Weakly Supervised Semantic Segmentation by Pixel-to-Prototype Contrast
2022.05.09
발표자: 윤수현
발표일자: 2022-05-09
저자: Ye Du, Zehua Fu, Qingjie Liu*, Yunhong Wang
학회명: CVPR 2022
Topological Attention for Time Series Forecasting
2022.05.09
발표자: 신효정
발표일자: 2022-05-09
저자: Sebastian Zeng, Florian Graf, Christoph Hofer, Roland Kwitt
학회명: NeurIPS 2021