English
Related papers

Related papers: Reinterpreting Importance-Weighted Autoencoders

200 papers

Variable importance is one of the most widely used measures for interpreting machine learning with significant interest from both statistics and machine learning communities. Recently, increasing attention has been directed toward…

Machine Learning · Statistics 2025-12-22 Xiaohan Wang , Yunzhe Zhou , Giles Hooker

We analyze variational inference for highly symmetric graphical models such as those arising from first-order probabilistic models. We first show that for these graphical models, the tree-reweighted variational objective lends itself to a…

Artificial Intelligence · Computer Science 2014-06-23 Hung Hai Bui , Tuyen N. Huynh , David Sontag

We develop a framework for incorporating structured graphical models in the \emph{encoders} of variational autoencoders (VAEs) that allows us to induce interpretable representations through approximate variational inference. This allows us…

The performance of sentence encoders can be significantly improved through the simple practice of fine-tuning using contrastive loss. A natural question arises: what characteristics do models acquire during contrastive learning? This paper…

Computation and Language · Computer Science 2023-10-25 Hiroto Kurita , Goro Kobayashi , Sho Yokoi , Kentaro Inui

We present a practical and statistically consistent scheme for actively learning binary classifiers under general loss functions. Our algorithm uses importance weighting to correct sampling bias, and by controlling the variance, we are able…

Machine Learning · Computer Science 2009-05-20 Alina Beygelzimer , Sanjoy Dasgupta , John Langford

A wide variety of model explanation approaches have been proposed in recent years, all guided by very different rationales and heuristics. In this paper, we take a new route and cast interpretability as a statistical inference problem. We…

Machine Learning · Computer Science 2024-01-01 Hugo Henri Joseph Senetaire , Damien Garreau , Jes Frellsen , Pierre-Alexandre Mattei

Powerful generative models, particularly in Natural Language Modelling, are commonly trained by maximizing a variational lower bound on the data log likelihood. These models often suffer from poor use of their latent variable, with ad-hoc…

Machine Learning · Statistics 2018-06-13 Alex Mansbridge , Roberto Fierimonte , Ilya Feige , David Barber

Interpretability is an important aspect of the trustworthiness of a model's predictions. Transformer's predictions are widely explained by the attention weights, i.e., a probability distribution generated at its self-attention unit (head).…

Computation and Language · Computer Science 2021-06-03 Rishabh Bhardwaj , Navonil Majumder , Soujanya Poria , Eduard Hovy

Autoencoders are unsupervised machine learning circuits whose learning goal is to minimize a distortion measure between inputs and outputs. Linear autoencoders can be defined over any field and only real-valued linear autoencoder have been…

Neural and Evolutionary Computing · Computer Science 2014-03-19 Pierre Baldi , Zhiqin Lu

Autoencoders receive latent models of input data. It was shown in recent works that they also estimate probability density functions of the input. This fact makes using the Bayesian decision theory possible. If we obtain latent models of…

Computer Vision and Pattern Recognition · Computer Science 2018-11-07 Vasily Morzhakov

Training deep generative models with maximum likelihood remains a challenge. The typical workaround is to use variational inference (VI) and maximize a lower bound to the log marginal likelihood of the data. Variational auto-encoders (VAEs)…

Machine Learning · Statistics 2019-08-13 Adji B. Dieng , John Paisley

Interpretability is essential in medical imaging to ensure that clinicians can comprehend and trust artificial intelligence models. Several approaches have been recently considered to encode attributes in the latent space to enhance its…

Image and Video Processing · Electrical Eng. & Systems 2024-08-27 Maxime Di Folco , Cosmin I. Bercea , Emily Chan , Julia A. Schnabel

Importance sampling is a widely used technique to estimate properties of a distribution. This paper investigates trading-off some bias for variance by adaptively winsorizing the importance sampling estimator. The novel winsorizing…

Computation · Statistics 2021-02-10 Paulo Orenstein

Importance sampling is often used in machine learning when training and testing data come from different distributions. In this paper we propose a new variant of importance sampling that can reduce the variance of importance sampling-based…

Machine Learning · Computer Science 2016-11-11 Philip S. Thomas , Emma Brunskill

Behavioural economics provides labels for patterns in human economic behaviour. Probability weighting is one such label. It expresses a mismatch between probabilities used in a formal model of a decision (i.e. model parameters) and…

Theoretical Economics · Economics 2020-05-04 Ole Peters , Alexander Adamou , Mark Kirstein , Yonatan Berman

Mechanistic interpretation has greatly contributed to a more detailed understanding of generative language models, enabling significant progress in identifying structures that implement key behaviors through interactions between internal…

Information Retrieval · Computer Science 2025-11-25 Meng Lu , Catherine Chen , Carsten Eickhoff

Motivated by distinct, though related, criteria, a growing number of attribution methods have been developed tointerprete deep learning. While each relies on the interpretability of the concept of "importance" and our ability to visualize…

Artificial Intelligence · Computer Science 2020-04-07 Zifan Wang , Piotr Mardziel , Anupam Datta , Matt Fredrikson

Importance sampling is widely used in machine learning and statistics, but its power is limited by the restriction of using simple proposals for which the importance weights can be tractably calculated. We address this problem by studying…

Machine Learning · Statistics 2016-10-18 Qiang Liu , Jason D. Lee

Interpretable deep learning is a fundamental building block towards safer AI, especially when the deployment possibilities of deep learning-based computer-aided medical diagnostic systems are so eminent. However, without a computational…

Machine Learning · Computer Science 2018-06-27 Anirban Mukhopadhyay

A probability distribution allows practitioners to uncover hidden structure in the data and build models to solve supervised learning problems using limited data. The focus of this report is on Variational autoencoders, a method to learn…

Machine Learning · Computer Science 2022-06-22 Vasanth Kalingeri