Designing Perceptual Puzzles by Differentiating Probabilistic Programs
arXiv:2204.12301 · doi:10.1145/3528233.3530715
Abstract
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.
arXiv comment9 pages; 3 figures; SIGGRAPH '22 Conference Proceedings
References in corpus (9)
- Auto-Encoding Variational Bayes
- Intriguing properties of neural networks
- Gradient-based Hyperparameter Optimization through Reversible Learning
- Markov Chain Monte Carlo and Variational Inference: Bridging the Gap
- Why Do Line Drawings Work? A Realism Hypothesis
- Approximate Bayesian Image Interpretation using Generative Probabilistic Graphics Programs
- Improved Contrastive Divergence Training of Energy Based Models
- Exponential Family Estimation via Adversarial Dynamics Embedding
- LF-PPL: A Low-Level First Order Probabilistic Programming Language for Non-Differentiable Models