QuBD extends algorithmic complexity estimation to quantized DNN weights, revealing that complexity decreases during learning, increases with overfitting, follows grokking patterns, and correlates with generalization.
Title resolution pending
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.LG 2years
2026 2verdicts
UNVERDICTED 2roles
background 1polarities
background 1representative citing papers
FedeKD adds energy-based sample-wise trust weighting to federated knowledge distillation, allowing proxy models to contribute more to reliable samples and less to unreliable ones, which reduces negative transfer in heterogeneous settings.
citing papers explorer
-
Characterizing Learning in Deep Neural Networks using Tractable Algorithmic Complexity Analysis
QuBD extends algorithmic complexity estimation to quantized DNN weights, revealing that complexity decreases during learning, increases with overfitting, follows grokking patterns, and correlates with generalization.
-
FedeKD: Energy-Based Gating for Robust Federated Knowledge Distillation under Heterogeneous Settings
FedeKD adds energy-based sample-wise trust weighting to federated knowledge distillation, allowing proxy models to contribute more to reliable samples and less to unreliable ones, which reduces negative transfer in heterogeneous settings.