mlsanityData to 7 Oct 2026

EBM Life Cycle: MCMC Strategies for Synthesis, Defense, and Density Modeling

arXiv 2205.12243 ↗First posted 24 May 2022stat.ML cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

This work presents strategies to learn an Energy-Based Model (EBM) according to the desired length of its MCMC sampling trajectories. MCMC trajectories of different lengths correspond to models with different purposes. Our experiments cover three different trajectory magnitudes and learning outcomes: 1) shortrun sampling for image generation; 2) midrun sampling for classifier-agnostic adversarial defense; and 3) longrun sampling for principled modeling of image probability densities. To achieve these outcomes, we introduce three novel methods of MCMC initialization for negative samples used in Maximum Likelihood (ML) learning. With standard network architectures and an unaltered ML objective, our MCMC initialization methods alone enable significant performance gains across the three applications that we investigate. Our results include state-of-the-art FID scores for unnormalized image densities on the CIFAR-10 and ImageNet datasets; state-of-the-art adversarial defense on CIFAR-10 among purification methods and the first EBM defense on ImageNet; and scalable techniques for learning valid probability densities. Code for this project can be found at https://github.com/point0bar1/ebm-life-cycle.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 47 of its 58 references that are in the map. First posted on arXiv 24 May 2022.

Builds on

47 of 58 references.

References · 47

  1. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  2. 2019EfficientNet: Rethinking Model Scaling for Convolutional Neural Networkscited by 6,415
  3. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
  4. 2022Generative Adversarial Networkscited by 5,603
  5. 2019Generative Modeling by Estimating Gradients of the Data Distributioncited by 4,067
  6. 2013Auto-Encoding Variational Bayescited by 3,980
  7. 2016Wide Residual Networkscited by 3,057
  8. 2018Large Scale GAN Training for High Fidelity Natural Image Synthesiscited by 2,609
  9. 2013Estimating or Propagating Gradients Through Stochastic Neurons for Conditional Computationcited by 2,031
  10. 2018Spectral Normalization for Generative Adversarial Networkscited by 1,618
  11. 2018Glow: Generative Flow with Invertible 1x1 Convolutionscited by 1,539
  12. 2019Theoretically Principled Trade-off between Robustness and Accuracycited by 1,510
35 more
  1. 2019Certified Adversarial Robustness via Randomized Smoothingcited by 1,324
  2. 2018Obfuscated Gradients Give a False Sense of Security: Circumventing Defenses to Adversarial Examplescited by 1,259
  3. 2020Improved Techniques for Training Score-Based Generative Modelscited by 1,036
  4. 2016Conditional Image Generation with PixelCNN Decoderscited by 859
  5. 2020Fast is better than free: Revisiting adversarial trainingcited by 723
  6. 2019Adversarial Training for Free!cited by 638
  7. 2020On Adaptive Attacks to Adversarial Example Defensescited by 547
  8. 2017Countering Adversarial Images using Input Transformationscited by 517
  9. 2019Unlabeled Data Improves Adversarial Robustnesscited by 415
  10. 2019Your Classifier is Secretly an Energy Based Model and You Should Treat it Like Onecited by 402
  11. 2015A note on the evaluation of generative modelscited by 291
  12. 2017PixelDefend: Leveraging Generative Models to Understand and Defend against Adversarial Examplescited by 243
  13. 2016A Theory of Generative ConvNetcited by 174
  14. 2019Are Labels Required for Improving Adversarial Robustness?cited by 162
  15. 2019Learning Non-Convergent Non-Persistent Short-Run MCMC Toward Energy-Based Modelcited by 147
  16. 2019Adversarial Robustness through Local Linearizationcited by 146
  17. 2021Adversarial purification with Score-based generative modelscited by 132
  18. 2020Improved Contrastive Divergence Training of Energy Based Modelscited by 127
  19. 2020Learning Energy-Based Models by Diffusion Recovery Likelihoodcited by 122
  20. 2019On the Anatomy of MCMC-Based Maximum Likelihood Learning of Energy-Based Modelscited by 105
  21. 2020VAEBM: A Symbiosis between Variational Autoencoders and Energy-based Modelscited by 103
  22. 2018Self-Supervised GANs via Auxiliary Rotation Losscited by 94
  23. 2019Flow Contrastive Estimation of Energy-Based Modelscited by 75
  24. 2020Your GAN is Secretly an Energy-based Model and You Should use Discriminator Driven Latent Samplingcited by 71
  25. 2019ME-Net: Towards Effective Adversarial Robustness with Matrix Estimationcited by 65
  26. 2020No MCMC for me: Amortized sampling for fast and stable training of energy-based modelscited by 64
  27. 2020Stochastic Security: Adversarial Defense Using Long-Run Dynamics of Energy-Based Modelscited by 58
  28. 2020InfoMax-GAN: Improved Adversarial Image Generation via Information Maximization and Contrastive Learningcited by 37
  29. 2016Deep Directed Generative Models with Energy-Based Probability Estimationcited by 34
  30. 2018Divergence Triangle for Joint Training of Generator Model, Energy-based Model, and Inference Modelcited by 34
  31. 2017Learning Energy-Based Models as Generative ConvNets via Multi-grid Modeling and Samplingcited by 32
  32. 2020Training Deep Energy-Based Models with f-Divergence Minimizationcited by 29
  33. 2021Equilibrium and non-Equilibrium regimes in the learning of Restricted Boltzmann Machinescited by 27
  34. 2017Wasserstein Introspective Neural Networkscited by 18
  35. 2020Mimicry: Towards the Reproducibility of GAN Researchcited by 14

Built on by

Not cited yet.