English
Related papers

Related papers: Score-Based Training for Energy-Based TTS Models

200 papers

This paper studies a family of estimators based on noise-contrastive estimation (NCE) for learning unnormalized distributions. The main contribution of this work is to provide a unified perspective on various methods for learning…

Machine Learning · Statistics 2025-07-15 J. Jon Ryu , Abhin Shah , Gregory W. Wornell

Energy-based models (EBMs) have become increasingly popular within computer vision in recent years. While they are commonly employed for generative image modeling, recent work has applied EBMs also for regression tasks, achieving…

Computer Vision and Pattern Recognition · Computer Science 2020-08-17 Fredrik K. Gustafsson , Martin Danelljan , Radu Timofte , Thomas B. Schön

In this work, we explore joint energy-based model (EBM) training during the finetuning of pretrained text encoders (e.g., Roberta) for natural language understanding (NLU) tasks. Our experiments show that EBM training can help the model…

Computation and Language · Computer Science 2021-02-22 Tianxing He , Bryan McCann , Caiming Xiong , Ehsan Hosseini-Asl

A crucial design decision for any robot learning pipeline is the choice of policy representation: what type of model should be used to generate the next set of robot actions? Owing to the inherent multi-modal nature of many robotic tasks,…

Robotics · Computer Science 2023-09-13 Sumeet Singh , Stephen Tu , Vikas Sindhwani

Estimating the parameters of probabilistic models of language such as maxent models and probabilistic neural models is computationally difficult since it involves evaluating partition functions by summing over an entire vocabulary, which…

Machine Learning · Computer Science 2014-10-31 Chris Dyer

Energy-Based Models (EBMs), also known as non-normalized probabilistic models, specify probability density or mass functions up to an unknown normalizing constant. Unlike most other probabilistic models, EBMs do not place a restriction on…

Machine Learning · Computer Science 2021-02-19 Yang Song , Diederik P. Kingma

The Energy-Based Model (EBM) framework is a very general approach to generative modeling that tries to learn and exploit probability distributions only defined though unnormalized scores. It has risen in popularity recently thanks to the…

Machine Learning · Computer Science 2022-02-25 Léo Gagnon , Guillaume Lajoie

By using the underlying theory of proper scoring rules, we design a family of noise-contrastive estimation (NCE) methods that are tractable for latent variable models. Both terms in the underlying NCE loss, the one using data samples and…

Machine Learning · Computer Science 2023-04-06 Christopher Zach

Contrastive learning has been demonstrated to be effective in enhancing pre-trained language models (PLMs) to derive superior universal sentence embeddings. However, existing contrastive methods still have two limitations. Firstly, previous…

Computation and Language · Computer Science 2022-10-11 Yuxin Jiang , Linhan Zhang , Wei Wang

We present a novel method for training score-based generative models which uses nonlinear noising dynamics to improve learning of structured distributions. Generalizing to a nonlinear drift allows for additional structure to be incorporated…

Machine Learning · Statistics 2025-07-10 Jeremiah Birrell , Markos A. Katsoulakis , Luc Rey-Bellet , Benjamin J. Zhang , Wei Zhu

Energy-Based Models (EBMs) offer a versatile framework for modeling complex data distributions. However, training and sampling from EBMs continue to pose significant challenges. The widely-used Denoising Score Matching (DSM) method for…

Machine Learning · Statistics 2024-03-20 Mingtian Zhang , Alex Hawkins-Hooker , Brooks Paige , David Barber

Score matching (SM) provides a compelling approach to learn energy-based models (EBMs) by avoiding the calculation of partition function. However, it remains largely open to learn energy-based latent variable models (EBLVMs), except some…

Machine Learning · Computer Science 2020-10-19 Fan Bao , Chongxuan Li , Kun Xu , Hang Su , Jun Zhu , Bo Zhang

Unnormalised latent variable models are a broad and flexible class of statistical models. However, learning their parameters from data is intractable, and few estimation techniques are currently available for such models. To increase the…

Machine Learning · Statistics 2019-02-26 Benjamin Rhodes , Michael Gutmann

Conditional Neural Processes~(CNPs) formulate distributions over functions and generate function observations with exact conditional likelihoods. CNPs, however, have limited expressivity for high-dimensional observations, since their…

Machine Learning · Computer Science 2023-03-24 Zesheng Ye , Jing Du , Lina Yao

Energy-based models (EBMs) are powerful probabilistic models, but suffer from intractable sampling and density evaluation due to the partition function. As a result, inference in EBMs relies on approximate sampling algorithms, leading to a…

Machine Learning · Computer Science 2020-01-10 Dieterich Lawson , George Tucker , Bo Dai , Rajesh Ranganath

Recent methods for learning unsupervised visual representations, dubbed contrastive learning, optimize the noise-contrastive estimation (NCE) bound on mutual information between two views of an image. NCE uses randomly sampled negative…

Machine Learning · Computer Science 2020-10-06 Mike Wu , Milan Mosse , Chengxu Zhuang , Daniel Yamins , Noah Goodman

A statistical model is said to be un-normalised when its likelihood function involves an intractable normalising constant. Two popular methods for parameter inference for these models are MC-MLE (Monte Carlo maximum likelihood estimation),…

Statistics Theory · Mathematics 2018-02-01 Lionel Riou-Durand , Nicolas Chopin

Learning a parametric model of a data distribution is a well-known statistical problem that has seen renewed interest as it is brought to scale in deep learning. Framing the problem as a self-supervised task, where data samples are…

Machine Learning · Statistics 2022-07-27 Omar Chehab , Alexandre Gramfort , Aapo Hyvarinen

A new whole-sentence language model - neural trans-dimensional random field language model (neural TRF LM), where sentences are modeled as a collection of random fields, and the potential function is defined by a neural network, has been…

Computation and Language · Computer Science 2018-07-04 Bin Wang , Zhijian Ou

Inferring the probability distribution of sentences or word sequences is a key process in natural language processing. While word-level language models (LMs) have been widely adopted for computing the joint probabilities of word sequences,…

Computation and Language · Computer Science 2021-03-16 Heewoong Park , Sukhyun Cho , Jonghun Park