REVIEW 6 cited by
Decoupled Kullback-Leibler Divergence Loss
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
Signed reviews
abstract
In this paper, we delve deeper into the Kullback-Leibler (KL) Divergence loss and mathematically prove that it is equivalent to the Decoupled Kullback-Leibler (DKL) Divergence loss that consists of 1) a weighted Mean Square Error (wMSE) loss and 2) a Cross-Entropy loss incorporating soft labels. Thanks to the decomposed formulation of DKL loss, we have identified two areas for improvement. Firstly, we address the limitation of KL/DKL in scenarios like knowledge distillation by breaking its asymmetric optimization property. This modification ensures that the $\mathbf{w}$MSE component is always effective during training, providing extra constructive cues. Secondly, we introduce class-wise global information into KL/DKL to mitigate bias from individual samples. With these two enhancements, we derive the Improved Kullback-Leibler (IKL) Divergence loss and evaluate its effectiveness by conducting experiments on CIFAR-10/100 and ImageNet datasets, focusing on adversarial training, and knowledge distillation tasks. The proposed approach achieves new state-of-the-art adversarial robustness on the public leaderboard -- RobustBench and competitive performance on knowledge distillation, demonstrating the substantial practical merits. Our code is available at https://github.com/jiequancui/DKL.
Forward citations
Cited by 6 Pith papers
-
Implicit Inversion turns CLIP into a Decoder
A frozen CLIP encoder can be inverted through frequency-aware implicit neural representations to generate images, reconstruct photos, edit scenes, and transfer style, all without training a decoder.
-
HEM: a margin-based loss for visual categorisation tasks
A new margin-based loss, HEM, trains image classifiers that are more robust to unknown and adversarial inputs and better at continual learning and segmentation than cross-entropy-trained models.
-
Towards Million-Scale Adversarial Robustness Evaluation With Stronger Individual Attacks
PMA, a probability-margin loss attack, consistently outperforms existing individual white-box attacks, and a one-million-image evaluation shows much lower robust accuracy than small-scale tests.
-
Wasserstein distributional adversarial training for deep neural networks
Extends TRADES to a Wasserstein distributional threat with an efficient budgeted attack, and fine-tunes pre-trained robust models to improve a self-defined distributional robustness metric.
-
Never Reset Again: A Mathematical Framework for Continual Inference in Recurrent Neural Networks
A dual cross-entropy and KL-divergence loss lets recurrent networks maintain stable accuracy over very long streams without hidden-state resets, matching and sometimes slightly beating periodic reset baselines.
-
Understanding Adversarial Training with Energy-based Models
Delta energy, the energy gap between an image and its adversarial counterpart, separates catastrophic from robust overfitting, and penalizing it with the DER regularizer mitigates both while improving generation diversity.
Discussion (0). Continue with ORCID to comment.