paper-with-me

홈 › Papers

VAE Approximation Error: ELBO and Exponential Families

2021-02-18 · ICLR 2022 4 · Alexander Shekhovtsov, Dmitrij Schlesinger, Boris Flach

The importance of Variational Autoencoders reaches far beyond standalone generative models -- the approach is also used for learning latent representations and can be generalized to semi-supervised learning. This requires a thorough analysis of their commonly known shortcomings: posterior collapse and approximation errors. This paper analyzes VAE approximation errors caused by the combination of the ELBO objective and encoder models from conditional exponential families, including, but not limited to, commonly used conditionally independent discrete and continuous models. We characterize subclasses of generative models consistent with these encoder families. We show that the ELBO optimizer is pulled away from the likelihood optimizer towards the consistent subset and study this effect experimentally. Importantly, this subset can not be enlarged, and the respective error cannot be decreased, by considering deeper encoder/decoder networks.

📄 PDF Abstract BibTeX arXiv:2102.09310

Code (0)

등록된 구현이 없습니다.

Tasks

Decoder

Methods 이 논문이 사용한 방법론

USD Coin Customer Service Number +1-833-534-1729 설명 없음

Similar Papers 제목 키워드 기반

SHOT-VAE: Semi-supervised Deep Generative Models With Label-aware ELBO Approximations

2020-11-21 · Hao-Zhe Feng, Kezhi Kong, Minghao Chen, Tianye Zhang 외

Semi-supervised variational autoencoders (VAEs) have obtained strong results, but have also encountered the challenge that good ELBO values do not always imply accurate inference results. In this paper, we investigate an…

4kSemi-Supervised Image ClassificationVariational Inference

Kernel Deformed Exponential Families for Sparse Continuous Attention

2021-11-01 · Alexander Moreno, Supriya Nagesh, Zhenke Wu, Walter Dempsey 외

Attention mechanisms take an expectation of a data representation with respect to probability weights. This creates summary statistics that focus on important features. Recently, (Martins et al. 2020, 2021) proposed cont…

Distributed Estimation, Information Loss and Exponential Families

2014-10-09 · NeurIPS 2014 12 · Qiang Liu, Alexander Ihler

Distributed learning of probabilistic models from multiple data repositories with minimum communication is increasingly important. We study a simple communication-efficient learning framework that first calculates the lo…

[Re] VAE Approximation Error: ELBO and Exponential Families

2023-09-21 · NeurIPS 2023 11

Scope of Reproducibility — Exponential family variational autoencoders struggle with reconstruction when encoders output limited information. We reproduce two experiments in which we first train the decoder and encoder s…

Batch and match: black-box variational inference with a score-based divergence

2024-02-22 · Diana Cai, Chirag Modi, Loucas Pillaud-Vivien, Charles C. Margossian 외

Most leading implementations of black-box variational inference (BBVI) are based on optimizing a stochastic evidence lower bound (ELBO). But such approaches to BBVI often converge slowly due to the high variance of their…

Variational Inference