Discrete Variational Autoencoding via Policy Search

• URL: http://arxiv.org/abs/2509.24716v1
• Date: Mon, 29 Sep 2025 12:44:05 GMT
• Title: Discrete Variational Autoencoding via Policy Search
• Abstract summary: discrete latent bottlenecks in variational autoencoders (VAEs) offer high bit efficiency.<n> discrete random variables do not allow for exact differentiable parameterization.<n>We propose a training framework for discrete VAEs that leverages the natural gradient of a non-parametric encoder.<n>Our method, combined with automatic step size adaptation and a transformer-based encoder, scales to challenging datasets such as ImageNet.
• Score: 16.257957838291563
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Discrete latent bottlenecks in variational autoencoders (VAEs) offer high bit efficiency and can be modeled with autoregressive discrete distributions, enabling parameter-efficient multimodal search with transformers. However, discrete random variables do not allow for exact differentiable parameterization; therefore, discrete VAEs typically rely on approximations, such as Gumbel-Softmax reparameterization or straight-through gradient estimates, or employ high-variance gradient-free methods such as REINFORCE that have had limited success on high-dimensional tasks such as image reconstruction. Inspired by popular techniques in policy search, we propose a training framework for discrete VAEs that leverages the natural gradient of a non-parametric encoder to update the parametric encoder without requiring reparameterization. Our method, combined with automatic step size adaptation and a transformer-based encoder, scales to challenging datasets such as ImageNet and outperforms both approximate reparameterization methods and quantization-based discrete autoencoders in reconstructing high-dimensional data from compact latent spaces, achieving a 20% improvement on FID Score for ImageNet 256.

Related papers

• Efficient Real-Time Adaptation of ROMs for Unsteady Flows Using Data Assimilation [7.958594167693376]
  We propose an efficient retraining strategy for a parameterized Reduced Order Model (ROM)<n>The strategy attains accuracy comparable to full retraining while requiring only a fraction of the computational time.<n>We show that, for the dynamical system considered, the dominant source of error in out-of-sample forecasts stems from distortions of the latent manifold.
  arXiv  Detail & Related papers  (2026-02-26T16:43:28Z)
• Element-wise Modulation of Random Matrices for Efficient Neural Layers [0.0]
  We propose a novel approach that decouples feature mixing from adaptation by utilizing a fixed random matrix modulated by lightweight, learnable element-wise parameters.<n>This architecture drastically reduces the trainable parameter count to a linear scale while retaining reliable accuracy across various benchmarks.
  arXiv  Detail & Related papers  (2025-12-15T16:16:53Z)
• Rethinking Autoregressive Models for Lossless Image Compression via Hierarchical Parallelism and Progressive Adaptation [75.58269386927076]
  Autoregressive (AR) models are often dismissed as impractical due to prohibitive computational cost.<n>This work re-thinks this paradigm, introducing a framework built on hierarchical parallelism and progressive adaptation.<n> Experiments on diverse datasets (natural, satellite, medical) validate that our method achieves new state-of-the-art compression.
  arXiv  Detail & Related papers  (2025-11-14T06:27:58Z)
• You Only Train Once [11.97836331714694]
  You Only Train Once (YOTO) contributes to limiting training to one shot for the latter aspect of losses selection and weighting.<n>We leverage the differentiability of the composite loss formulation which is widely used for optimizing multiple empirical losses simultaneously.<n>We show that YOTO consistently outperforms the best grid-search model on unseen test data.
  arXiv  Detail & Related papers  (2025-06-04T18:04:58Z)
• Likelihood-Free Variational Autoencoders [21.092166159353702]
  Variational Autoencoders (VAEs) typically rely on a probabilistic decoder with a predefined likelihood, most commonly an isotropic Gaussian, to model the data conditional on latent variables.<n>We propose EnVAE, a novel likelihood-free generative framework that has a deterministic decoder and employs the energy score--a proper scoring rule--to build the reconstruction loss.<n>Our framework offers a general, scalable, and statistically principled alternative for flexible and nonparametric distribution learning in generative modeling.
  arXiv  Detail & Related papers  (2025-04-24T14:44:46Z)
• Multi-Scale Invertible Neural Network for Wide-Range Variable-Rate Learned Image Compression [90.59962443790593]
  In this paper, we present a variable-rate image compression model based on invertible transform to overcome limitations.<n> Specifically, we design a lightweight multi-scale invertible neural network, which maps the input image into multi-scale latent representations.<n> Experimental results demonstrate that the proposed method achieves state-of-the-art performance compared to existing variable-rate methods.
  arXiv  Detail & Related papers  (2025-03-27T09:08:39Z)
• Reduced Order Modeling with Shallow Recurrent Decoder Networks [5.686433280542813]
  SHRED-ROM is a robust decoding-only strategy that encodes the numerically unstable approximation of an inverse.<n>We show that SHRED-ROM accurately reconstructs the state dynamics for new parameter values starting from limited fixed or mobile sensors.
  arXiv  Detail & Related papers  (2025-02-15T23:41:31Z)
• ALoRE: Efficient Visual Adaptation via Aggregating Low Rank Experts [71.91042186338163]
  ALoRE is a novel PETL method that reuses the hypercomplex parameterized space constructed by Kronecker product to Aggregate Low Rank Experts.<n>Thanks to the artful design, ALoRE maintains negligible extra parameters and can be effortlessly merged into the frozen backbone.
  arXiv  Detail & Related papers  (2024-12-11T12:31:30Z)
• Ensemble Kalman Filtering Meets Gaussian Process SSM for Non-Mean-Field and Online Inference [47.460898983429374]
  We introduce an ensemble Kalman filter (EnKF) into the non-mean-field (NMF) variational inference framework to approximate the posterior distribution of the latent states. This novel marriage between EnKF and GPSSM not only eliminates the need for extensive parameterization in learning variational distributions, but also enables an interpretable, closed-form approximation of the evidence lower bound (ELBO) We demonstrate that the resulting EnKF-aided online algorithm embodies a principled objective function by ensuring data-fitting accuracy while incorporating model regularizations to mitigate overfitting.
  arXiv  Detail & Related papers  (2023-12-10T15:22:30Z)
• Patch Similarity Aware Data-Free Quantization for Vision Transformers [2.954890575035673]
  We propose PSAQ-ViT, a Patch Similarity Aware data-free Quantization framework for Vision Transformers. We analyze the self-attention module's properties and reveal a general difference (patch similarity) in its processing of Gaussian noise and real images. Experiments and ablation studies are conducted on various benchmarks to validate the effectiveness of PSAQ-ViT.
  arXiv  Detail & Related papers  (2022-03-04T11:47:20Z)
• Rate Distortion Characteristic Modeling for Neural Image Compression [59.25700168404325]
  End-to-end optimization capability offers neural image compression (NIC) superior lossy compression performance. distinct models are required to be trained to reach different points in the rate-distortion (R-D) space. We make efforts to formulate the essential mathematical functions to describe the R-D behavior of NIC using deep network and statistical modeling.
  arXiv  Detail & Related papers  (2021-06-24T12:23:05Z)
• On the Encoder-Decoder Incompatibility in Variational Text Modeling and Beyond [82.18770740564642]
  Variational autoencoders (VAEs) combine latent variables with amortized variational inference. We observe the encoder-decoder incompatibility that leads to poor parameterizations of the data manifold. We propose Coupled-VAE, which couples a VAE model with a deterministic autoencoder with the same structure.
  arXiv  Detail & Related papers  (2020-04-20T10:34:10Z)
• Implicit differentiation of Lasso-type models for hyperparameter optimization [82.73138686390514]
  We introduce an efficient implicit differentiation algorithm, without matrix inversion, tailored for Lasso-type problems. Our approach scales to high-dimensional data by leveraging the sparsity of the solutions.
  arXiv  Detail & Related papers  (2020-02-20T18:43:42Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.