Modelpedia
Work in progress
About
Findings
Models
Concepts
Methods
Datasets
Sources
Light
Dark
Faithful and Efficient Explanations for Neural Networks via Neural Tangent Kernel Surrogate Models
2024-01-16
· ICLR 2024 spotlight ·
anchor
Findings
IC-1602
ResNet18, ResNet34, and MobileNetV2 pre-trained on CIFAR10 have decision functions well-approximated by a kernel machine using the trace NTK, with Kendall-τ correlations of 0.776, 0.786, and 0.700
IC-1603
ResNet18's CIFAR10 classification decisions are driven by the bulk of training data rather than a sparse set of exemplars, as revealed by trntk data attribution