Learning Robust Global Representations by Penalizing Local ...haohanw/PAR/poster.pdfLearning Robust...
Transcript of Learning Robust Global Representations by Penalizing Local ...haohanw/PAR/poster.pdfLearning Robust...
Learning Robust Global Representations by Penalizing Local Predictive Power Haohan Wang, Songwei Ge, Eric P. Xing, Zachary C. Lipton
School of Computer Science, Carnegie Mellon University
! ImageNet-Sketch Dataset & Experiments
• First out-of-domain data set at the ImageNet validation set scale • 1000 classes, with 50 testing images in each • Used as test data set to test the model’s generalization
ability when trained on standard ImageNet train set. • Performance:
• Analysis:
Accuracy AlexNet DANN* InfoDrop HEX PAR Top1 0.1204 0.1360* 0.1224 0.1292 0.1306 Top5 0.2480 0.2712* 0.2560 0.2654 0.2627
AlexNet-PAR AlexNet Predic;on Confidence Predic;on Confidence
stethoscope 0.6608 hook 0.3903
tricycle 0.9260 safetypin 0.5143
Afghanhound 0.8945 swab(mop) 0.7379
redwine 0.5999 goblet 0.7427
! Patch-wise Adversarial Regularization (PAR) ! Highlights
! Empirical Results
• Notations • top layers: f(•;θ) • patch classifier: h(•;ϕ) • bottom layers: g(•;δ)
• Patch-wise Adversarial Regularization
• Training heuristics • first train the model conventionally until
convergence • then train the model with regularization
• Variants • PAR: • 1-layer classifier • 1x1 local patch • first layer
• PARB • 3x3 local path
• PARM • 3-layer classifier
• PARH • higher layer
• Engineering-wise • One set of parameters • Implemented efficiently
through convolution
• Out-of-domain CIFAR10 • Test with ResNet-50 • 4 out-of-domain settings created: • Greyscale, NegativeColor,
RandomKernel, RadiamKernel • Best performance in comparison to
standard methods
• PACS experiment • Test with AlexNet (consistent with
previous state-of-the-art) • Best average performance in
domain-agnostic setting • Best performance in Sketch domain
in comparison to any method
! Contact
• Novel method for out-of-domain robustness • with domain-agnostic setting (more
industry-friendly) • simple and intuitive regularization,
architecture-agnostic • New vision data set for large scale out-of-
domain robustness testing • ImageNet validation set scale
! Motivation • Neural networks are not robust enough! • Models with high accuracy can easily fail
when tested with out-of-domain data • One reason is that the models are
exploiting predictive local signals, ignoring the global picture
• Penalize model’s tendency in predicting through local signals
• [email protected] @HaohanWang • [email protected] • resource links