paper-with-me

Papers

Bayesian Active Learning By Distribution Disagreement

2025-01-02 · Thorben Werner, Lars Schmidt-Thieme

Active Learning (AL) for regression has been systematically under-researched due to the increased difficulty of measuring uncertainty in regression models. Since normalizing flows offer a full predictive distribution instead of a point forecast, they facilitate direct usage of known heuristics for AL like Entropy or Least-Confident sampling. However, we show that most of these heuristics do not work well for normalizing flows in pool-based AL and we need more sophisticated algorithms to distinguish between aleatoric and epistemic uncertainty. In this work we propose BALSA, an adaptation of the BALD algorithm, tailored for regression with normalizing flows. With this work we extend current research on uncertainty quantification with normalizing flows \cite{berry2023normalizing, berry2023escaping} to real world data and pool-based AL with multiple acquisition functions and query sizes. We report SOTA results for BALSA across 4 different datasets and 2 different architectures.

📄 PDF Abstract BibTeX arXiv:2501.01248

Code (1)

wernerth94/bayesian-active-learning-by-distribution-disagreement 공식 구현 pytorch

Tasks

Active LearningregressionUncertainty Quantification

Methods 이 논문이 사용한 방법론

Normalizing Flows Normalizing Flows are a method for constructing complex distributions by transforming a probability density through a series of invertible mappings. By repeatedly applying…

Similar Papers 제목 키워드 기반

Bayesian Active Learning for Censored Regression

2024-02-19 · Frederik Boe Hüttel, Christoffer Riis, Filipe Rodrigues, Francisco Câmara Pereira

Bayesian active learning is based on information theoretical approaches that focus on maximising the information that new observations provide to the model parameters. This is commonly done by maximising the Bayesian Act…

Active Learningregression

Bayesian Active Learning by Disagreements: A Geometric Perspective

2021-05-06 · Xiaofeng Cao, Ivor W. Tsang

We present geometric Bayesian active learning by disagreements (GBALD), a framework that performs BALD on its core-set construction interacting with model uncertainty estimation. Technically, GBALD constructs core-set on…

Active Learning

Mitigating Sampling Bias and Improving Robustness in Active Learning

2021-09-13 · Ranganath Krishnan, Alok Sinha, Nilesh Ahuja, Mahesh Subedar 외

This paper presents simple and efficient methods to mitigate sampling bias in active learning while achieving state-of-the-art accuracy and model robustness. We introduce supervised contrastive active learning by leverag…

Active Learning

Self-Correcting Bayesian Optimization through Bayesian Active Learning

2023-04-21 · NeurIPS 2023 11 · Carl Hvarfner, Erik Hellsten, Frank Hutter, Luigi Nardi

Gaussian processes are the model of choice in Bayesian optimization and active learning. Yet, they are highly dependent on cleverly chosen hyperparameters to reach their full potential, and little effort is devoted to fi…

Active LearningBayesian OptimizationGaussian Processes

Batch Bayesian Active Learning with Partial Batch Label Sampling

2025-10-10 · Kangping Hu, Stephen Mussmann arxiv

Over the past couple of decades, many active learning acquisition functions have been proposed, leaving practitioners with an unclear choice of which to use. Bayesian-based active learning offers principled objectives wi…

Active Learning