analiza aiqbench
This research introduces a training-adaptive framework for convolutional sparse coding that treats the sparsity coefficient as a learnable variable within an information bottleneck perspective. By jointly optimizing this parameter with network weights, the method achieves robust visual representations on CIFAR and ImageNet while maintaining competitive accuracy on clean data. The inclusion of a label-free post-training strategy further enhances resilience against input perturbations without requiring additional model retraining.
Training-Adaptive Convolutional Sparse Coding via Information Bottleneck for Robust Visual Representation
arXiv:2609.19122v2 Announce Type: cross Abstract: 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.
Abstract: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.
Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG); Machine Learning (stat.ML)
arXiv:2609.19122 [cs.CV]
(or arXiv:2609.19122v2 [cs.CV] for this version)
https://doi.org/10.48550/arXiv.2609.19122 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 250 cuvinte
Subiecte conexe
PIRAMID: Progress and Plans
LessWrong · 27 iul. 2026
The Machines that Make the Machines
NVIDIA · 7 oct. 2026
Pathway’s brain-inspired architecture development on Amazon SageMaker HyperPod
AWS · 8 sept. 2026
Expenditure Horizon: Measuring Optimization Ability, with an Application to NanoGPT
METR · 21 iul. 2026