HOLE applies persistent homology to latent embeddings in neural networks and uses visualizations such as cluster flow diagrams to reveal patterns of class separation, feature disentanglement, and robustness.
Zeiler and Rob Fergus
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.LG 2years
2025 2verdicts
UNVERDICTED 2representative citing papers
Flipping 1-2 sign bits in DNN parameters, located without data or optimization, drops accuracy to near zero across image classification, detection, segmentation, and language models.
citing papers explorer
-
HOLE: Homological Observation of Latent Embeddings for Neural Network Interpretability
HOLE applies persistent homology to latent embeddings in neural networks and uses visualizations such as cluster flow diagrams to reveal patterns of class separation, feature disentanglement, and robustness.
-
Maximal Brain Damage Without Data or Optimization: Disrupting Neural Networks via Sign-Bit Flips
Flipping 1-2 sign bits in DNN parameters, located without data or optimization, drops accuracy to near zero across image classification, detection, segmentation, and language models.