MLSP-L20: Representation Learning and Model Design in Deep Networks
Oral
Thu, 7 May, 09:00 - 11:00
Location: Room 117
Session Type: Oral
Track: Machine Learning for Signal Processing [ML]
Click the to view the manuscript on IEEE Xplore Open Preview
Thu, 7 May, 09:00 - 09:20

MLSP-L20.1: Prompt-Guided Multi-Scale Feature Pyramid Aggregation with Unified Channel-Spatial Transformer for Single Image Deraining
Haoshen Li, Zifan Chen, Guangzhengao Yang, Jie Zhao, Bin Dong, Li Zhang, Peking University, China
Thu, 7 May, 09:20 - 09:40

MLSP-L20.2: ZEN-DARTS: MITIGATING PERFORMANCE COLLAPSE WITH SYNFLOW METRIC REGULARIZATION AND IMPROVED ARCHITECTURE PARAMETER INITIALIZATION
Boyu Wang, Yong Zhong, Chengdu Institute of Computer Applications, Chinese Academy of Sciences/University of Chinese Academy of Sciences, China; Yang Liu, Shanghai Jiao Tong University, China
Thu, 7 May, 09:40 - 10:00

MLSP-L20.3: SMEKGE: A SHRINKAGE-GUIDED META-ENSEMBLE OF KNOWLEDGE GRAPH EMBEDDING EXPERTS
Linjia Hao, Yongli Hu, Tong Bie, Huajie Jiang, Kan Guo, Tengfei Liu, Beijing University of Technology, China; Junbin Gao, The University of Sydney, China; Yanfeng Sun, Baocai Yin, Beijing University of Technology, China
Thu, 7 May, 10:00 - 10:20

MLSP-L20.4: LEARNABLE INSTANCE ATTENTION FILTERING FOR ADAPTIVE DETECTOR DISTILLATION
Chen Liu, Qizhen Lan, University of Alabama at Birmingham, United States of America; Zhicheng Ding, Xinyu Chu, Bowling Green State University, United States of America; Qing Tian, University of Alabama at Birmingham, United States of America
Thu, 7 May, 10:20 - 10:40

MLSP-L20.5: DEEPTRAVERSE: AN ALGORITHM-INSPIRED DESIGN PARADIGM FOR STRUCTURED AND INTERPRETABLE VISION BACKBONES
Bin Guo, John H.L. Hansen, University of Texas at Dallas, United States of America
Thu, 7 May, 10:40 - 11:00

MLSP-L20.6: T-GEMS: TEXT-GUIDED EXIT MODULES FOR DECREASING CLIP IMAGE ENCODER
Alberto Presta, Grzegorz Stefanski, Samsung AI Center Warsaw, Poland; Michal Byra, Samsung AI Center Warsaw, Institute of Fundamental Technological Research, PAS, Warsaw, Poland; Krzysztof Arendt, Samsung AI Center Warsaw, Poland