paper-with-me

홈 › Papers

On the Convergence of Loss and Uncertainty-based Active Learning Algorithms

2023-12-21 · Daniel Haimovich, Dima Karamshuk, Fridolin Linder, Niek Tax, Milan Vojnovic

We investigate the convergence rates and data sample sizes required for training a machine learning model using a stochastic gradient descent (SGD) algorithm, where data points are sampled based on either their loss value or uncertainty value. These training methods are particularly relevant for active learning and data subset selection problems. For SGD with a constant step size update, we present convergence results for linear classifiers and linearly separable datasets using squared hinge loss and similar training loss functions. Additionally, we extend our analysis to more general classifiers and datasets, considering a wide range of loss-based sampling strategies and smooth convex training loss functions. We propose a novel algorithm called Adaptive-Weight Sampling (AWS) that utilizes SGD with an adaptive step size that achieves stochastic Polyak's step size in expectation. We establish convergence rate results for AWS for smooth convex training loss functions. Our numerical experiments demonstrate the efficiency of AWS on various datasets by using either exact or estimated loss values.

📄 PDF Abstract BibTeX arXiv:2312.13927

Code (1)

facebookresearch/adaptiveweightsampling 공식 구현 tf

Tasks

Active Learning

Methods 이 논문이 사용한 방법론

SGD Stochastic Gradient Descent is an iterative optimization technique that uses minibatches of data to form an expectation of the gradient, rather than the full gradient using…

Similar Papers 제목 키워드 기반

Convergence of Uncertainty Sampling for Active Learning

2021-10-29 · Anant Raj, Francis Bach

Uncertainty sampling in active learning is heavily used in practice to reduce the annotation cost. However, there has been no wide consensus on the function to be used for uncertainty estimation in binary classification …

Active LearningBinary ClassificationClassificationMulti-class Classification

Understanding Uncertainty Sampling

2023-07-06 · Shang Liu, Xiaocheng Li

Uncertainty sampling is a prevalent active learning algorithm that queries sequentially the annotations of data samples which the current prediction model is uncertain about. However, the usage of uncertainty sampling ha…

Active Learning

Online Active Learning of Reject Option Classifiers

2019-06-14 · Kulin Shah, Naresh Manwani

Active learning is an important technique to reduce the number of labeled examples in supervised learning. Active learning for binary classification has been well addressed in machine learning. However, active learning o…

Active LearningBinary ClassificationGeneral Classification

Improving greedy core-set configurations for active learning with uncertainty-scaled distances

2022-02-09 · Yuchen Li, Frank Rudzicz

We scale perceived distances of the core-set algorithm by a factor of uncertainty and search for low-confidence configurations, finding significant improvements in sample efficiency across CIFAR10/100 and SVHN image clas…

Active Learningimage-classificationImage Classification

Uncertainty-aware Active Learning for Optimal Bayesian Classifier

2021-01-01 · ICLR 2021 1 · Guang Zhao, Edward Dougherty, Byung-Jun Yoon, Francis Alexander 외

For pool-based active learning, in each iteration a candidate training sample is chosen for labeling by optimizing an acquisition function. Expected Loss Reduction~(ELR) methods maximize the expected reduction in the cla…

Active LearningClassificationGeneral Classification