Training-Adaptive Convolutional Sparse Coding via Information Bottleneck for Robust Visual Representation
Paper proposes training-adaptive convolutional sparse coding with a learnable sparsity coefficient for robust visual representations.
The framework unrolls convolutional sparse coding optimization with FISTA and treats the sparsity coefficient as a differentiable variable learned jointly with network parameters, interpreted via the information bottleneck. A label-free post-training strategy adjusts compression strength for corrupted inputs with the main network frozen. Experiments on CIFAR and ImageNet show competitive clean accuracy and greatly improved robustness under input perturbations.
- Sparsity coefficient learned as differentiable variable via FISTA unrolling
- Information bottleneck view balances compression and signal retention
- Label-free post-training adapts compression for corrupted inputs
- Validated on CIFAR and ImageNet with improved perturbation robustness
Full article149 words · extracted from huggingface.co · click to collapse
Visual signals require compact yet sufficient representations for robust downstream prediction. Convolutional sparse coding (CSC) provides an explicit mechanism for suppressing redundant components while preserving signal content, but its sparsity coefficient is typically fixed and manually selected. We propose a training-adaptive convolutional sparse coding framework for robust visual signal representation. Specifically, we unfold the CSC optimization with the Fast Iterative Shrinkage-Thresholding Algorithm (FISTA) and treat the sparsity coefficient as a differentiable variable jointly learned with the network parameters. From the information bottleneck perspective, this coefficient controls the trade-off between information retention and compression: the sparsity term promotes compact representations, while the reconstruction term together with task loss preserves task-relevant signal content. We further introduce a label-free post-training strategy that adjusts the compression strength for corrupted inputs with the main network parameters fixed. Experiments on CIFAR and ImageNet demonstrate competitive clean-data recognition and greatly improved robustness under different input perturbations.
Text extracted automatically; images, tables and formatting may be missing. Original: https://huggingface.co/papers/2609.19122