activity
20242026
collaborators

8 papers

cs.LG2026

DiffGradCAM: A Class Activation Map Using the Full Model Decision to Solve Unaddressed Adversarial Attacks

Jacob Piland, Chris Sweet, Adam Czajka

Class Activation Mapping (CAM) and its gradient-based variants (e.g., GradCAM) have become standard tools for explaining Convolutional Neural Network (CNN) predictions. However, th…

cs.CV2025

SAGE: Saliency-Guided Contrastive Embeddings

Colton R. Crum, Christopher Sweet, Adam Czajka

Integrating human perceptual priors into the training of neural networks has been shown to raise model generalization, serve as an effective regularizer, and align models with huma…

cs.CV2025

Almost Right: Making First-Layer Kernels Nearly Orthogonal Improves Model Generalization

Colton R. Crum, Adam Czajka

Despite several algorithmic advances in the training of convolutional neural networks (CNNs) over the years, their generalization capabilities are still subpar across several perti…

cs.CV2025

Saliency-Guided Training for Fingerprint Presentation Attack Detection

Samuel Webster, Adam Czajka

Saliency-guided training, which directs model learning to important regions of images, has demonstrated generalization improvements across various biometric presentation attack det…

cs.CV2025

Divisive Decisions: Improving Salience-Based Training for Generalization in Binary Classification Tasks

Jacob Piland, Chris Sweet, Adam Czajka

Existing saliency-guided training approaches improve model generalization by incorporating a loss term that compares the model's class activation map (CAM) for a sample's true-clas…

cs.CV2025

MENTOR: Human Perception-Guided Pretraining for Increased Generalization

Colton R. Crum, Adam Czajka

Leveraging human perception into training of convolutional neural networks (CNN) has boosted generalization capabilities of such models in open-set recognition tasks. One of the ac…