FP-V1.V5.11
EFFICIENT FINE-TUNING OF DEEP NEURAL NETWORKS WITH EFFECTIVE PARAMETER ALLOCATION
Phillip Wallis, Oregon Health & Science University / Microsoft, United States of America; Xubo Song, Oregon Health & Science University, United States of America
Session:
New Techniques for Learning
Track:
Applications of Machine Learning
Location:
Gather.Town 5
Presentation Time:
Fri, 7 Oct, 21:00 - 22:00 China Standard Time (UTC +8)
Fri, 7 Oct, 15:00 - 16:00 Central European Time (UTC +1)
Fri, 7 Oct, 13:00 - 14:00 UTC
Fri, 7 Oct, 09:00 - 10:00 Eastern Time (UTC -5)
Fri, 7 Oct, 15:00 - 16:00 Central European Time (UTC +1)
Fri, 7 Oct, 13:00 - 14:00 UTC
Fri, 7 Oct, 09:00 - 10:00 Eastern Time (UTC -5)
Session Co-Chairs:
Jean-Christophe Pesquet, CentraleSupélec and Andrea Cavallaro, Queen Mary University of London and Rebecca Willett, University of Chicago
Presentation
Discussion
Resources
No resources available.
Session FP-V1.V5
FP-V1.V5.1: TRUNCATED LOTTERY TICKET FOR DEEP PRUNING
Iraj Saniee, Lisa Zhang, Bell Labs, Nokia, United States of America; Bradley Magnetta, Yale University, United States of America
FP-V1.V5.2: Which Metrics For Network Pruning: Final Accuracy? or Accuracy Drop?
Donggyu Joo, Sunghyun Baek, Junmo Kim, KAIST, Korea, Republic of
FP-V1.V5.3: Improving Generalization of Reinforcement Learning using a Bilinear Policy Network
Fen Fang, Wenyu Liang, Yan Wu, Qianli Xu, Joo-hwee Lim, Home address, Singapore
FP-V1.V5.4: RAPID: A SINGLE STAGE PRUNING FRAMEWORK
Ankit Sharma, Hassan Foroosh, University of Central Florida, United States of America
FP-V1.V5.5: RETHINKING EFFICACY OF SOFTMAX FOR LIGHTWEIGHT NON-LOCAL NEURAL NETWORKS
Yooshin Cho, Youngsoo Kim, Hanbyel Cho, Jaesung Ahn, Hyeong Gwon Hong, Junmo Kim, Korea Advanced Institute of Science and Technology, Korea, Republic of
FP-V1.V5.6: Two distillation perspectives based on Tanimoto coefficient
Hongqiao Shu, Lenovo, China
FP-V1.V5.7: D-CBRS: ACCOUNTING FOR INTRA-CLASS DIVERSITY IN CONTINUAL LEARNING
Yasin Findik, Farhad Pourkamali-Anaraki, University of Massachusetts Lowell, United States of America
FP-V1.V5.8: META-BNS FOR ADVERSARIAL DATA-FREE QUANTIZATION
Siming Fu, Hualiang Wang, Yuchen Cao, Haoji Hu, Zhejiang University, China; Bo Peng, Wenming Tan, Tingqun Ye, Hikvision Research Institute, China
FP-V1.V5.9: DEEP RESIDUAL NETWORKS WITH COMMON LINEAR MULTI-STEP AND ADVANCED NUMERICAL SCHEMES
Zhengbo Luo, Weilian Zhou, Sei-ichiro Kamata, Waseda University, Japan; Xuehui Hu, Tongtai Information Technology Co., Ltd., China
FP-V1.V5.10: A LOW-COMPLEXITY MODIFIED THINET ALGORITHM FOR PRUNING CONVOLUTIONAL NEURAL NETWORKS
Sadegh Tofigh, M. Omair Ahmad, M.N.S Swamy, Concordia University, Canada
FP-V1.V5.11: EFFICIENT FINE-TUNING OF DEEP NEURAL NETWORKS WITH EFFECTIVE PARAMETER ALLOCATION
Phillip Wallis, Oregon Health & Science University / Microsoft, United States of America; Xubo Song, Oregon Health & Science University, United States of America
FP-V1.V5.12: ADAPTIVE PROXY ANCHOR LOSS FOR DEEP METRIC LEARNING
Nguyen Phan, Sen Tran, Ta Duc Huy, Steven Q.H. Truong, VinBrain, Viet Nam; Soan T.M. Duong, VinBrain, Le Quy Don Technical University, Viet Nam; Chanh D.Tr. Nguyen, VinBrain, VinUniversity, Viet Nam; Trung Bui, No affiliation, Viet Nam
FP-V1.V5.13: Latent Preserving Generative Adversarial Network for Imbalance classification
Tanmoy Dam, Sreenatha G. Anavatti, Hussein A. Abbass, University of New South Wales, Canberra, Australia; Md Meftahul Ferdaus, Nanyang Technological University Singapore, Singapore; Mahardhika Pratama, University of South Australia,, Australia; Senthilnath Jayavelu, Institute for Infocomm Research, Singapore
FP-V1.V5.14: MULTI-STEP TEST-TIME ADAPTATION WITH ENTROPY MINIMIZATION AND PSEUDO-LABELING
Hiroaki Kingetsu, Kenichi Kobayashi, Yoshihiro Okawa, Yasuto Yokota, Katsuhito Nakazawa, AI Laboratory, Fujitsu Limited, Japan
FP-V1.V5.15: Occlusion-invariant Representation Alignment for Entity Re-identification
Zhanghao Jiang, Ke Xu, Heshan Du, Huan Jin, Zheng Lu, Qian Zhang, University of Nottingham Ningbo China, China