Towards interpreting deep neural networks via layer behavior understanding

DOI10.1007/s10994-021-06074-8zbMath1491.68178OpenAlexW4220673036WikidataQ114955313 ScholiaQ114955313MaRDI QIDQ2673336

Mingkui Tan, Xiping Hu, Xiangmiao Wu, Jiezhang Cao, Jincheng Li

Publication date: 9 June 2022

Published in: Machine Learning (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1007/s10994-021-06074-8

zbMATH Keywords

Wasserstein distancelayer behaviorteacher-student paradigm

Mathematics Subject Classification ID

Artificial neural networks and deep learning (68T07) Optimal transportation (49Q22)

Uses Software

• word2vec
• ImageNet
• CNN-RNN
• SVCCA
• PASCAL VOC
• BigGAN

Cites Work

• Unnamed Item
• Unnamed Item
• Computational Optimal Transport: With Applications to Data Science
• Gradient descent optimizes over-parameterized deep ReLU networks
• Few-shot learning with adaptively initialized task optimizer: a practical meta-learning approach
• Spanning attack: reinforce black-box attacks with unlabeled data
• The Sinkhorn–Knopp Algorithm: Convergence and Applications
• On the information bottleneck theory of deep learning
• Dynamics of stochastic gradient descent for two-layer neural networks in the teacher–student setup*
• Optimal Transport