Towards interpreting deep neural networks via layer behavior understanding

DOI10.1007/S10994-021-06074-8MaRDI QIDQ2673336zbMATH OpenOpenAlexWikidataFDO

Authors Jiezhang Cao, Jin-cheng Li, Xiping Hu, Xiangmiao Wu, Mingkui Tan

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

layer behaviorteacher-student paradigmWasserstein distance

Mathematics Subject Classification ID

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

Recommendations

• Transport analysis of infinitely deep neural network
• Understanding Deep Learning with Statistical Relevance
• scientific article; zbMATH DE number 6610876
• Mean Field Analysis of Deep Neural Networks
• Archetypal landscapes for deep neural networks

Cites work

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

Cited in

(4)

• Interpreting deep learning models with marginal attribution by conditioning on quantiles
• scientific article; zbMATH DE number 6610876 (Why is no real title available?)
• Transport analysis of infinitely deep neural network
• Progressive Interpretation Synthesis: Interpreting Task Solving by Quantifying Previously Used and Unused Information

Describes a project that uses

Uses Software

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

This page was built for publication: Towards interpreting deep neural networks via layer behavior understanding

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2673336)