Inverting Deep Generative models, One layer at a time

by   Qi Lei, et al.

We study the problem of inverting a deep generative model with ReLU activations. Inversion corresponds to finding a latent code vector that explains observed measurements as much as possible. In most prior works this is performed by attempting to solve a non-convex optimization problem involving the generator. In this paper we obtain several novel theoretical results for the inversion problem. We show that for the realizable case, single layer inversion can be performed exactly in polynomial time, by solving a linear program. Further, we show that for multiple layers, inversion is NP-hard and the pre-image set can be non-convex. For generative models of arbitrary depth, we show that exact recovery is possible in polynomial time with high probability, if the layers are expanding and the weights are randomly selected. Very recent work analyzed the same problem for gradient descent inversion. Their analysis requires significantly higher expansion (logarithmic in the latent dimension) while our proposed algorithm can provably reconstruct even with constant factor expansion. We also provide provable error bounds for different norms for reconstructing noisy observations. Our empirical validation demonstrates that we obtain better reconstructions when the latent dimension is large.


page 8

page 9

page 16


Neural Spectrahedra and Semidefinite Lifts: Global Convex Optimization of Polynomial Activation Neural Networks in Fully Polynomial-Time

The training of two-layer neural networks with nonlinear activation func...

I Don't Need 𝐮: Identifiable Non-Linear ICA Without Side Information

In this work we introduce a new approach for identifiable non-linear ICA...

A Generalization of Wirtinger Flow for Exact Interferometric Inversion

Interferometric inversion involves recovery of a signal from cross-corre...

Regularized Training of Intermediate Layers for Generative Models for Inverse Problems

Generative Adversarial Networks (GANs) have been shown to be powerful an...

On the Fine-Grained Hardness of Inverting Generative Models

The objective of generative model inversion is to identify a size-n late...

Stochastic seismic waveform inversion using generative adversarial networks as a geological prior

We present an application of deep generative models in the context of pa...

Gradient-Free Textual Inversion

Recent works on personalized text-to-image generation usually learn to b...

Please sign up or login with your details

Forgot password? Click here to reset