Understanding Adversarial Training with Energy-based Models

• URL: http://arxiv.org/abs/2505.22486v1
• Date: Wed, 28 May 2025 15:36:02 GMT
• Title: Understanding Adversarial Training with Energy-based Models
• Authors: Mujtaba Hussain Mirza, Maria Rosaria Briglia, Filippo Bartolucci, Senad Beadini, Giuseppe Lisanti, Iacopo Masi,
• Abstract summary: We aim at using Energy-based Model (EBM) framework to better understand adversarial training (AT) in classifiers.<n>We analyze the impact of existing AT approaches on the energy of samples during training.<n>We propose a novel regularizer, the Delta Energy Regularizer (DER), designed to smoothen the energy landscape during training.
• Score: 4.147437142690764
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: We aim at using Energy-based Model (EBM) framework to better understand adversarial training (AT) in classifiers, and additionally to analyze the intrinsic generative capabilities of robust classifiers. By viewing standard classifiers through an energy lens, we begin by analyzing how the energies of adversarial examples, generated by various attacks, differ from those of the natural samples. The central focus of our work is to understand the critical phenomena of Catastrophic Overfitting (CO) and Robust Overfitting (RO) in AT from an energy perspective. We analyze the impact of existing AT approaches on the energy of samples during training and observe that the behavior of the ``delta energy' -- change in energy between original sample and its adversarial counterpart -- diverges significantly when CO or RO occurs. After a thorough analysis of these energy dynamics and their relationship with overfitting, we propose a novel regularizer, the Delta Energy Regularizer (DER), designed to smoothen the energy landscape during training. We demonstrate that DER is effective in mitigating both CO and RO across multiple benchmarks. We further show that robust classifiers, when being used as generative models, have limits in handling trade-off between image quality and variability. We propose an improved technique based on a local class-wise principal component analysis (PCA) and energy-based guidance for better class-specific initialization and adaptive stopping, enhancing sample diversity and generation quality. Considering that we do not explicitly train for generative modeling, we achieve a competitive Inception Score (IS) and Fr\'echet inception distance (FID) compared to hybrid discriminative-generative models.

Related papers

• Your Classifier Can Do More: Towards Bridging the Gaps in Classification, Robustness, and Generation [18.149950949071982]
  We study the energy distribution differences of clean, adversarial, and generated samples across various JEM variants and adversarially trained models.<n>We propose Energy-based Joint Distribution Adrialversa Training to jointly model the clean data distribution, the adversarial distribution, and the classifier.
  arXiv  Detail & Related papers  (2025-05-26T03:26:55Z)
• Shedding More Light on Robust Classifiers under the lens of Energy-based Models [3.953603590878949]
  We offer a new take on the dynamics of adversarial training (AT) Our analysis of the energy landscape during AT reveals that untargeted attacks generate adversarial images much more in-distribution (lower energy) than the original data from the point of view of the model. Motivated by rigorous evidence, we propose Weighted Energy Adversarial Training (WEAT)
  arXiv  Detail & Related papers  (2024-07-08T18:31:19Z)
• Revisiting Energy Based Models as Policies: Ranking Noise Contrastive Estimation and Interpolating Energy Models [18.949193683555237]
  In this work, we revisit the choice of energy-based models (EBM) as a policy class. We develop a training objective and algorithm for energy models which combines several key ingredients. We show that the Implicit Behavior Cloning (IBC) objective is actually biased even at the population level.
  arXiv  Detail & Related papers  (2023-09-11T20:13:47Z)
• Learning Energy-Based Models by Cooperative Diffusion Recovery Likelihood [64.95663299945171]
  Training energy-based models (EBMs) on high-dimensional data can be both challenging and time-consuming. There exists a noticeable gap in sample quality between EBMs and other generative frameworks like GANs and diffusion models. We propose cooperative diffusion recovery likelihood (CDRL), an effective approach to tractably learn and sample from a series of EBMs.
  arXiv  Detail & Related papers  (2023-09-10T22:05:24Z)
• On Feature Diversity in Energy-based Models [98.78384185493624]
  An energy-based model (EBM) is typically formed of inner-model(s) that learn a combination of the different features to generate an energy mapping for each input configuration. We extend the probably approximately correct (PAC) theory of EBMs and analyze the effect of redundancy reduction on the performance of EBMs.
  arXiv  Detail & Related papers  (2023-06-02T12:30:42Z)
• Latent Diffusion Energy-Based Model for Interpretable Text Modeling [104.85356157724372]
  We introduce a novel symbiosis between the diffusion models and latent space EBMs in a variational learning framework. We develop a geometric clustering-based regularization jointly with the information bottleneck to further improve the quality of the learned latent space.
  arXiv  Detail & Related papers  (2022-06-13T03:41:31Z)
• A Unified Contrastive Energy-based Model for Understanding the Generative Ability of Adversarial Training [64.71254710803368]
  Adversarial Training (AT) is an effective approach to enhance the robustness of deep neural networks. We demystify this phenomenon by developing a unified probabilistic framework, called Contrastive Energy-based Models (CEM) We propose a principled method to develop adversarial learning and sampling methods.
  arXiv  Detail & Related papers  (2022-03-25T05:33:34Z)
• Energy-based View of Retrosynthesis [70.66156081030766]
  We propose a framework that unifies sequence- and graph-based methods as energy-based models. We present a novel dual variant within the framework that performs consistent training over Bayesian forward- and backward-prediction. This model improves state-of-the-art performance by 9.6% for template-free approaches where the reaction type is unknown.
  arXiv  Detail & Related papers  (2020-07-14T18:51:06Z)
• Joint Training of Variational Auto-Encoder and Latent Energy-Based Model [112.7509497792616]
  This paper proposes a joint training method to learn both the variational auto-encoder (VAE) and the latent energy-based model (EBM) The joint training of VAE and latent EBM are based on an objective function that consists of three Kullback-Leibler divergences between three joint distributions on the latent vector and the image.
  arXiv  Detail & Related papers  (2020-06-10T20:32:25Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.