mlsanityData to 7 Oct 2026

Designing Perceptual Puzzles by Differentiating Probabilistic Programs

arXiv 2204.12301 ↗First posted 26 Apr 2022cs.GR cs.AI cs.LGcited by 12

Not on the map. Why?

Abstract (arXiv)

We design new visual illusions by finding "adversarial examples" for principled models of human perception -- specifically, for probabilistic models, which treat vision as Bayesian inference. To perform this search efficiently, we design a differentiable probabilistic programming language, whose API exposes MCMC inference as a first-class differentiable function. We demonstrate our method by automatically creating illusions for three features of human vision: color constancy, size constancy, and face perception.

Key facts. Cited by 12 papers in the map (data through 7 Oct 2026). Builds on 19 of its 65 references that are in the map. First posted on arXiv 26 Apr 2022.

Builds on

19 of 65 references.

References · 19

  1. 2013Intriguing properties of neural networkscited by 5,114
  2. 2013Auto-Encoding Variational Bayescited by 3,980
  3. 2019Soft Rasterizer: A Differentiable Renderer for Image-based 3D Reasoningcited by 420
  4. 2015Gradient-based Hyperparameter Optimization through Reversible Learningcited by 357
  5. 2019Monte Carlo Gradient Estimation in Machine Learningcited by 281
  6. 2017Morphable Face Models - An Open Frameworkcited by 138
  7. 2020Improved Contrastive Divergence Training of Energy Based Modelscited by 127
  8. 2018Adversarial Examples that Fool both Computer Vision and Time-Limited Humanscited by 73
  9. 2018An Introduction to Probabilistic Programmingcited by 62
  10. 2021Differentiable Annealed Importance Sampling and the Perils of Gradient Noisecited by 36
  11. 2018Humans can decipher adversarial imagescited by 31
  12. 2019Exponential Family Estimation via Adversarial Dynamics Embeddingcited by 30
7 more
  1. 2020Why Do Line Drawings Work? A Realism Hypothesiscited by 28
  2. 20213DP3: 3D Scene Perception via Probabilistic Programmingcited by 26
  3. 2015A New Approach to Probabilistic Programming Inferencecited by 22
  4. 2013Approximate Bayesian Image Interpretation using Generative Probabilistic Graphics Programscited by 10
  5. 2019LF-PPL: A Low-Level First Order Probabilistic Programming Language for Non-Differentiable Modelscited by 3
  6. 2019Reducing Lateral Visual Biases in Displayscited by 2
  7. 2019Disagreeing about Crocs and socks: Creating profoundly ambiguous color displayscited by 1

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 5; 2024: 4; 2025: 2; 2026: 1.

12 papers.

  1. 2023Visual Anagrams: Generating Multi-View Optical Illusions with Diffusion Modelscited by 46
  2. 2024Factorized Diffusion: Perceptual Illusions by Noise Decompositioncited by 23
  3. 2025LookingGlass: Generative Anamorphoses via Laplacian Pyramid Warpingcited by 7
  4. 2024Diffusion-based Visual Anagram as Multi-task Learningcited by 6
  5. 2023Acting as Inverse Inverse Planningcited by 5
  6. 2023Differentiating Metropolis-Hastings to Optimize Intractable Densitiescited by 5
  7. 2023Zero Grads: Learning Local Surrogate Losses for Non-Differentiable Graphicscited by 5
  8. 2024Illusion3D: 3D Multiview Illusion with 2D Diffusion Priorscited by 5
  9. 2023Inferring the Future by Imagining the Pastcited by 4
  10. 2025Hand-Shadow Posercited by 3
  11. 2024Sketching With Your Voice: "Non-Phonorealistic" Rendering of Sounds via Vocal Imitationcited by 1
  12. 2026Meschers: Geometry Processing of Impossible Objectscited by 1