Implicit Discriminator in Variational Autoencoder

  title={Implicit Discriminator in Variational Autoencoder},
  author={Prateek Munjal and Akanksha Paul and N. C. Krishnan},
  journal={2020 International Joint Conference on Neural Networks (IJCNN)},
Recently generative models have focused on combining the advantages of variational autoencoders (VAE) and generative adversarial networks (GAN) for good reconstruction and generative abilities. In this work we introduce a novel hybrid architecture, Implicit Discriminator in Variational Autoencoder (IDVAE), that combines a VAE and a GAN, which does not need an explicit discriminator network. The fundamental premise of the IDVAE architecture is that the encoder of a VAE and the discriminator of a… Expand
An adversarial algorithm for variational inference with a new role for acetylcholine
This work constructs a VI system that is both compatible with neurobiology and avoids the assumption that neural activities are independent given lower-layers during generation, and implements this algorithm, which can successfully train the approximate inference network for generative models. Expand
Bio-informed Protein Sequence Generation for Multi-class Virus Mutation Prediction
A GAN-based multi-class protein sequence generative model, named ProteinSeqGAN, which can be promising predictions of virus mutations and can generate valid antigen protein sequences from both bioinformatics and statistics perspectives. Expand
Unsupervised Anomaly Detection with Adversarial Mirrored AutoEncoders
Adversarial Mirrored AutoEncoder is proposed, a simple modification to Adversarial Autoencoder where a Mirrored Wasserstein loss in the Discriminator is used to enforce better semantic-level reconstruction and a new metric for anomaly quantification is proposed. Expand


Variational Approaches for Auto-Encoding Generative Adversarial Networks
This paper develops a principle upon which auto-encoders can be combined with generative adversarial networks by exploiting the hierarchical structure of the generative model, and describes a unified objective for optimization. Expand
Adversarial Variational Bayes: Unifying Variational Autoencoders and Generative Adversarial Networks
Adversarial Variational Bayes (AVB), a technique for training Variational Autoencoders with arbitrarily expressive inference models by introducing an auxiliary discriminative network that allows to rephrase the maximum-likelihood-problem as a two-player game, hence establishing a principled connection between VAEs and Generative Adversarial Networks (GANs). Expand
Improved Training of Generative Adversarial Networks Using Representative Features
This paper achieves both aims simultaneously by improving the stability of training GANs by implicitly regularizing the discriminator using representative features based on the fact that standard GAN minimizes reverse Kullback-Leibler divergence. Expand
Adversarial Generator-Encoder Networks
We present a new autoencoder-type architecture, that is trainable in an unsupervised mode, sustains both generation and inference, and has the quality of conditional and unconditional samples boostedExpand
Improved Techniques for Training GANs
This work focuses on two applications of GANs: semi-supervised learning, and the generation of images that humans find visually realistic, and presents ImageNet samples with unprecedented resolution and shows that the methods enable the model to learn recognizable features of ImageNet classes. Expand
It Takes (Only) Two: Adversarial Generator-Encoder Networks
We present a new autoencoder-type architecture that is trainable in an unsupervised mode, sustains both generation and inference, and has the quality of conditional and unconditional samples boostedExpand
Mode Regularized Generative Adversarial Networks
This work introduces several ways of regularizing the objective, which can dramatically stabilize the training of GAN models and shows that these regularizers can help the fair distribution of probability mass across the modes of the data generating distribution, during the early phases of training and thus providing a unified solution to the missing modes problem. Expand
Adversarial Symmetric Variational Autoencoder
An extensive set of experiments is performed, in which a new form of adversarial training is developed and state-of-the-art data reconstruction and generation on several image benchmark datasets are demonstrated. Expand
Generative Adversarial Nets
We propose a new framework for estimating generative models via an adversarial process, in which we simultaneously train two models: a generative model G that captures the data distribution, and aExpand
Autoencoding beyond pixels using a learned similarity metric
An autoencoder that leverages learned representations to better measure similarities in data space is presented and it is shown that the method learns an embedding in which high-level abstract visual features (e.g. wearing glasses) can be modified using simple arithmetic. Expand