English
Related papers

Related papers: Masking schemes for universal marginalisers

200 papers

Mask-based pretraining has become a cornerstone of modern large-scale models across language, vision, and recently biology. Despite its empirical success, its role and limits in learning data representations have been unclear. In this work,…

Machine Learning · Computer Science 2025-09-29 Mingze Dong , Leda Wang , Yuval Kluger

We propose a simple, training-free mechanism which explains the generalization behaviour of diffusion models. By comparing pre-trained diffusion models to their theoretically optimal empirical counterparts, we identify a shared local…

Machine Learning · Computer Science 2025-06-11 Matthew Niedoba , Berend Zwartsenberg , Kevin Murphy , Frank Wood

Likelihood based-learning of graphical models faces challenges of computational-complexity and robustness to model mis-specification. This paper studies methods that fit parameters directly to maximize a measure of the accuracy of predicted…

Machine Learning · Computer Science 2014-07-04 Justin Domke

Since the development of self-supervised visual representation learning from contrastive learning to masked image modeling (MIM), there is no significant difference in essence, that is, how to design proper pretext tasks for vision…

Computer Vision and Pattern Recognition · Computer Science 2023-01-31 Kun Yi , Yixiao Ge , Xiaotong Li , Shusheng Yang , Dian Li , Jianping Wu , Ying Shan , Xiaohu Qie

In this work, we dive deep into the impact of additive noise in pre-training deep networks. While various methods have attempted to use additive noise inspired by the success of latent denoising diffusion models, when used in combination…

Computer Vision and Pattern Recognition · Computer Science 2024-12-30 Hyesong Choi , Daeun Kim , Sungmin Cha , Kwang Moo Yi , Dongbo Min

Neural solvers for partial differential equations (PDEs) have great potential to generate fast and accurate physics solutions, yet their practicality is currently limited by their generalizability. PDEs evolve over broad scales and exhibit…

Machine Learning · Computer Science 2024-12-06 Anthony Zhou , Amir Barati Farimani

Masked Diffusion Models (MDMs) have emerged as one of the most promising paradigms for generative modeling over discrete domains. It is known that MDMs effectively train to decode tokens in a random order, and that this ordering has…

Machine Learning · Computer Science 2025-11-25 Prateek Garg , Bhavya Kohli , Sunita Sarawagi

We develop an approach to training generative models based on unrolling a variational auto-encoder into a Markov chain, and shaping the chain's trajectories using a technique inspired by recent work in Approximate Bayesian computation. We…

Machine Learning · Computer Science 2017-08-03 Philip Bachman , Doina Precup

A significant effort has been made to train neural networks that replicate algorithmic reasoning, but they often fail to learn the abstract concepts underlying these algorithms. This is evidenced by their inability to generalize to data…

Machine Learning · Computer Science 2020-10-26 Yujun Yan , Kevin Swersky , Danai Koutra , Parthasarathy Ranganathan , Milad Hashemi

We study how masking and predicting tokens in an unsupervised fashion can give rise to linguistic structures and downstream performance gains. Recent theories have suggested that pretrained language models acquire useful inductive biases…

Computation and Language · Computer Science 2021-04-13 Tianyi Zhang , Tatsunori Hashimoto

Training neural network models with discrete (categorical or structured) latent variables can be computationally challenging, due to the need for marginalization over large or combinatorial sets. To circumvent this issue, one typically…

Machine Learning · Computer Science 2020-12-29 Gonçalo M. Correia , Vlad Niculae , Wilker Aziz , André F. T. Martins

With the advent of billion-parameter foundation models, efficient fine-tuning has become increasingly important for the adaptation of models to downstream tasks. However, especially in computer vision, it can be hard to achieve good…

Computer Vision and Pattern Recognition · Computer Science 2024-09-13 Alfonso Taboada Warmerdam , Mathilde Caron , Yuki M. Asano

The global inducing point variational approximation for BNNs is based on using a set of inducing inputs to construct a series of conditional distributions that accurately approximate the conditionals of the true posterior distribution. Our…

Machine Learning · Statistics 2023-10-25 Matthew Ashman , Tommy Rochussen , Adrian Weller

Self-supervised learning methods like masked autoencoders (MAE) have shown significant promise in learning robust feature representations, particularly in image reconstruction-based pretraining task. However, their performance is often…

Computer Vision and Pattern Recognition · Computer Science 2025-07-31 Sua Lee , Joonhun Lee , Myungjoo Kang

In this work, we observe that model trained on vast general images via masking strategy, has been naturally embedded with their distribution knowledge, thus spontaneously attains the underlying potential for strong image denoising. Based on…

Computer Vision and Pattern Recognition · Computer Science 2024-11-20 Xiaoxiao Ma , Zhixiang Wei , Yi Jin , Pengyang Ling , Tianle Liu , Ben Wang , Junkang Dai , Huaian Chen

We investigate the learning dynamics of classifiers in scenarios where classes are separable or classifiers are over-parameterized. In both cases, Empirical Risk Minimization (ERM) results in zero training error. However, there are many…

Machine Learning · Computer Science 2024-10-23 Julius Martinetz , Christoph Linse , Thomas Martinetz

Promising results have driven a recent surge of interest in continuous optimization methods for Bayesian network structure learning from observational data. However, there are theoretical limitations on the identifiability of underlying…

We introduce a novel training principle for probabilistic models that is an alternative to maximum likelihood. The proposed Generative Stochastic Networks (GSN) framework is based on learning the transition operator of a Markov chain whose…

Machine Learning · Computer Science 2015-03-29 Guillaume Alain , Yoshua Bengio , Li Yao , Jason Yosinski , Eric Thibodeau-Laufer , Saizheng Zhang , Pascal Vincent

Masked image modeling (MIM) has become a leading self-supervised learning strategy. MIMs such as Masked Autoencoder (MAE) learn strong representations by randomly masking input tokens for the encoder to process, with the decoder…

Computer Vision and Pattern Recognition · Computer Science 2024-08-27 Taekyung Kim , Sanghyuk Chun , Byeongho Heo , Dongyoon Han

The crucial role of convolutional models, both as standalone vision models and backbones in foundation models, necessitates effective acceleration techniques. This paper proposes a novel method to learn semi-structured sparsity patterns for…

Machine Learning · Computer Science 2024-11-04 David A. Danhofer