Research map / Label-efficient and robust learning / Recognition and long-tail
Visualizing Deep Neural Network Decisions: Prediction Difference Analysis
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use conditional sampling instead of marginal sampling (ablation) | Prediction Difference Analysis · ImageNet | explanation specificity | Helped |
| vary patch size: larger patches smooth the visualization, while very large patches make it blurry (ablation) | Prediction Difference Analysis with AlexNet · ImageNet | visualization interpretability | Mixed |
| use conditional sampling with single-pixel patches (k=1) (ablation) | Prediction Difference Analysis with AlexNet · ImageNet | effect on classifier prediction | Helped |
| condition on spatial grid location during conditional sampling (ablation) | Prediction Difference Analysis · COBRA MRI | interpretability | Helped |
| use prediction-difference visualizations instead of logistic-regression weights | MRI classifier explanations · COBRA MRI | visualization specificity | Helped |
Closest papers by meaning
- DiG-IN: Diffusion Guidance for Investigating Networks - Uncovering Classifier Differences Neuron Visualisations and Visual Counterfactual Explanations (CVPR 2024 · similarity 0.85)
- Don't trust your eyes: on the (un)reliability of feature visualizations (ICML 2024 · similarity 0.84)
- CAPE: CAM as a Probabilistic Ensemble for Enhanced DNN Interpretation (CVPR 2024 · similarity 0.84)
- Representational Similarity via Interpretable Visual Concepts (ICLR 2025 · similarity 0.83)
- Neuro-Visualizer: A Novel Auto-Encoder-Based Loss Landscape Visualization Method With an Application in Knowledge-Guided Machine Learning (ICML 2024 · similarity 0.83)
- Understanding Generalization through Decision Pattern Shift (NeurIPS 2026 · similarity 0.83)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.