Bayesian Reasoning with Trained Neural Networks
arXiv:2001.11031 · doi:10.3390/e23060693
Abstract
We showed how to use trained neural networks to perform Bayesian reasoning in order to solve tasks outside their initial scope. Deep generative models provide prior knowledge, and classification/regression networks impose constraints. The tasks at hand were formulated as Bayesian inference problems, which we approximately solved through variational or sampling techniques. The approach built on top of already trained networks, and the addressable questions grew super-exponentially with the number of available networks. In its simplest form, the approach yielded conditional generative models. However, multiple simultaneous constraints constitute elaborate questions. We compared the approach to specifically trained generators, showed how to solve riddles, and demonstrated its compatibility with state-of-the-art architectures.
References in corpus (18)
- Auto-Encoding Variational Bayes
- Conditional Generative Adversarial Nets
- Intriguing properties of neural networks
- GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash Equilibrium
- Variational Inference: A Review for Statisticians
- Deep Convolutional Neural Network for Inverse Problems in Imaging
- Generative Adversarial Text to Image Synthesis
- Conditional Image Generation with PixelCNN Decoders
- Improved Training of Wasserstein GANs
- Variational Inference with Normalizing Flows
- Compressed Sensing using Generative Models
- Variational Continual Learning
- Plug & Play Generative Networks: Conditional Iterative Generation of Images in Latent Space
- Generative Visual Manipulation on the Natural Image Manifold
- Precise Recovery of Latent Vectors from Generative Adversarial Networks
- Uncertainty Quantification with Generative Models
- Metric Gaussian Variational Inference
- Conditional Inference in Pre-trained Variational Autoencoders via Cross-coding