paper-with-me

Papers

$f$-Divergence Thermodynamic Variational Objective: a Deformed Geometry Perspective

2021-09-29 · Jun Li, Ping Li

In this paper, we propose a $f$-divergence Thermodynamic Variational Objective ($f$-TVO). $f$-TVO generalizes the Thermodynamic Variational Objective (TVO) by replacing Kullback–Leibler (KL) divergence with arbitary differeitiable $f$-divergence. In particular, $f$-TVO approximates dual function of model evidence $f^*(p(x))$ rather than the log model evidence $\log p(x)$ in TVO. $f$-TVO is derived from a deformed $\chi$-geometry perspective. By defining $\chi$-exponential family exponential, we are able to integral $f$-TVO along the $\chi$-path, which is the deformed geodesic between variational posterior distribution and true posterior distribution. Optimizing scheme of $f$-TVO includes reparameterization trick and Monte Carlo approximation. Experiments on VAE and Bayesian neural network show that the proposed $f$-TVO performs better than cooresponding baseline $f$-divergence variational inference.

📄 PDF Abstract BibTeX

Code (0)

등록된 구현이 없습니다.

Tasks

Variational Inference

Similar Papers 제목 키워드 기반

All in the (Exponential) Family: Information Geometry and Thermodynamic Variational Inference

2020-01-01 · ICML 2020 1 · Rob Brekelmans, Vaden Masrani, Frank Wood, Greg Ver Steeg 외

While the Evidence Lower Bound (ELBO) has become a ubiquitous objective for variational inference, the recently proposed Thermodynamic Variational Objective (TVO) leverages thermodynamic integration to provide a tighter …

AllSchedulingVariational Inference

Variational Inference with Holder Bounds

2021-11-04 · Junya Chen, Danni Lu, Zidi Xiu, Ke Bai 외

The recent introduction of thermodynamic integration techniques has provided a new framework for understanding and improving variational inference (VI). In this work, we present a careful analysis of the thermodynamic va…

Variational Inference

All in the Exponential Family: Bregman Duality in Thermodynamic Variational Inference

2020-07-01 · Rob Brekelmans, Vaden Masrani, Frank Wood, Greg Ver Steeg 외

The recently proposed Thermodynamic Variational Objective (TVO) leverages thermodynamic integration to provide a family of variational inference objectives, which both tighten and generalize the ubiquitous Evidence Lower…

AllVariational Inference

Quadratic polarity and polar Fenchel-Young divergences from the canonical Legendre polarity

2026-03-05 · Frank Nielsen, Basile Plus-Gourdon, Mahito Sugiyama arxiv

Polarity is a fundamental reciprocal duality of $n$-dimensional projective geometry which associates to points polar hyperplanes, and more generally $k$-dimensional convex bodies to polar $(n-1-k)$-dimensional convex bod…

The Thermodynamic Variational Objective

2019-06-28 · NeurIPS 2019 12 · Vaden Masrani, Tuan Anh Le, Frank Wood

We introduce the thermodynamic variational objective (TVO) for learning in both continuous and discrete deep generative models. The TVO arises from a key connection between variational inference and thermodynamic integra…

Variational Inference