Adversarially Regularized Autoencoders

By Junbo Jake Zhao, Yoon Kim, Kelly W. Zhang, Alexander M. Rush, Yann LeCun. 2017

300 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Generative Models

Citations in the corpus

Residual Energy-Based Models for Text Generation (2020, background): “Several variants of auto-encoders have also been investigated for representing and generating text (Bowman et al., 2016; Zhao et al., 2018), but they have not shown signiﬁcant improvements in terms of perplexity and they have so far been applied to relatively small datasets only.”
Diagnosing and Enhancing VAE Models (2019, methodology): “For example, only log-likelihood values on relatively small black-and-white images are reported in (Tomczak and Welling, 2018), while discrete data and associated evaluation metrics are the focus of (Zhao et al., 2018).”
Analysis Methods in Neural Language Processing: A Survey (2018, methodology): “A black-box targeted attack for MT was proposed by Zhao et al. (2018c), who used GANs to search for attacks on Google’s MT system after mapping sentences into continuous space with adversarially regularized autoen-coders (Zhao et al., 2018b).”
Learning Representations by Maximizing Mutual Information in Variational Autoencoders (2019, background): “Adversarial autoencoders (AAE) [13], and adversarial regularized autoencoders (ARAE) [14] are VAE-based methods proposed to leverage adversarial learning [15].”
Learning Multimodal Graph-to-Graph Translation for Molecular Optimization (2018, methodology): “Our adversarial training technique is inspired by recent text style transfer methods (Shen et al., 2017; Zhao et al., 2018) that adversarially regularize the continuous representation of discrete structures to enable end-to-end training.”
Continual Learning for Text Classification with Information Disentanglement Based Regularization (2021, background): “Textual Information Disentanglement Our work is related to information disentanglement for text data, which has been extensively explored in generation tasks like style transfer (Fu et al., 2017; Zhao et al., 2018; Romanov et al., 2018; Li et al., 2020), where text hidden representations are often d”
Out-of-Domain Detection for Natural Language Understanding in Dialog Systems (2019, background): “lable text generation [35]–[37], some of which also involve an adversarial training process [38], [39] and controls the generated content.”
Generating Natural Adversarial Examples (2017, methodology): “We use the adversarially regularized autoencoder (ARAE) [23] for encoding discrete text into continuous codes, and introduce an inverter that maps these continuous codes into the Gaussian space of z.”
Delete, Retrieve, Generate: a Simple Approach to Sentiment and Style Transfer (2018, methodology): “Most existing work (Shen et al., 2017; Zhao et al., 2018; Fu et al., 2018; Melnyk et al., 2017) uses adversarial training to separate attribute and content: the content encoder aims to fool the attribute discriminator by removing attribute information from the content embedding.”
Learning Latent Space Energy-Based Prior Model (2020, methodology): “Text datasets include PTB [40], Yahoo [61], and SNLI [6], following recent work on text generative modeling with latent variables [27, 63, 35].”
Generating Sentences from Disentangled Syntactic and Semantic Spaces (2019, methodology): “Clearly, DSS-VAE outperforms a plain VAE in BLEU if Forward PPL is controlled, and in Forward PPL if BLEU is controlled.”
Triple-to-Text: Converting RDF Triples into High-Quality Natural Languages via Optimizing an Inverse KL Divergence (2019, background): “[31], FPPL measures the fluency of generated sentences.”
Optimus: Organizing Sentences via Pre-trained Modeling of a Latent Space (2020, background): “Note our two-stage ﬁne-tuning scheme borrows the ideas from ARAE and NN-Outlines.”
Pretraining Text Encoders with Adversarial Mixture of Training Signal Generators (2022, background): “…the ablation studies of ELECTRA (Clark et al., 2020), or directly operating in the hidden states of the generator instead of in its discrete output space which has been studied in adversarial learning based text generation research (Subramanian et al., 2018; Zhang et al., 2017; Zhao et al., 2018).”
Stochastic Wasserstein Autoencoder for Probabilistic Sentence Generation (2018, methodology): “Alternatively, the dual form (adversarial loss) can also be used for WAE (Zhao et al., 2018).”
Residual Energy-Based Models for Text (2021, methodology): “However, most works have so far been applied to relatively small datasets and compared with small-to-medium-sized baseline models only (Bowman et al., 2016; Zhao et al., 2018; He et al., 2019).”
Deep Learning for Text Style Transfer: A Survey (2020, methodology): “AE is used in many TST works (e.g., Shen et al. 2017; Hu et al. 2017; Fu et al. 2018; Zhao et al. 2018; Prabhumoye et al. 2018; Yang et al. 2018).”