Active Learning

Active learning is a machine learning paradigm focused on optimizing data labeling efficiency by strategically selecting the most informative samples for annotation from a larger unlabeled pool. Current research emphasizes developing novel acquisition functions and data pruning strategies to reduce computational costs associated with large datasets, exploring the integration of active learning with various model architectures (including deep neural networks, Gaussian processes, and language models), and addressing challenges like privacy preservation and handling open-set noise. This approach holds significant promise for reducing the substantial cost and effort of data labeling in diverse fields, ranging from image classification and natural language processing to materials science and healthcare.

Papers

October 12, 2022

VL4Pose: Active Learning Through Out-Of-Distribution Detection For Pose Estimation
Megh Shukla, Roshan Roy, Pankaj Singh, Shuaib Ahmed, Alexandre Alahi
Active Learning Distribution Detection Pose Estimation Human Pose Device Learning Pose Refinement

October 11, 2022

Deep Active Ensemble Sampling For Image Classification
Salman Mohamadi, Gianfranco Doretto, Donald A. Adjeroh
Active Learning Image Classification Sample Selection Better Uncertainty Estimation Resource Efficient Deep Geometric Framework

October 9, 2022

Few-Shot Continual Active Learning by a Robot
Ali Ayub, Carter Fendley
Continual LEArning Active Learning Robot Person Object Classification Label Supervision Realistic Continual Learning

October 7, 2022

Is margin all you need? An extensive empirical study of active learning on tabular data
Dara Bahri, Heinrich Jiang, Tal Schuster, Afshin Rostamizadeh
Deep Neural Network Active Learning Empirical Study Tabular Data Margin Maximization Label Constraint

October 6, 2022

To Softmax, or not to Softmax: that is the question when applying Active Learning for Transformer Models
Julius Gonsior, Christian Falkenberg, Silvio Magino, Anja Reusch, Maik Thiele, Wolfgang Lehner
Active Learning Transformer Model Yes No Question Transformer Based Language Model Softmax Function Natural Language Processing Application

October 5, 2022

October 3, 2022

Nonstationary data stream classification with online active learning and siamese neural networks
Kleanthis Malialis, Christos G. Panayiotou, Marios M. Polycarpou
Classification Code Active Learning Siamese Network Novel Active Learning Non Stationary Data

October 2, 2022

Improved Algorithms for Neural Active Learning
Yikun Ban, Yuheng Zhang, Hanghang Tong, Arindam Banerjee, Jingrui He
Active Learning Multi Class Classification Instance Dependent Regret Improved Algorithm ActIve Learning algORithm Static Regret Neural Active Learning

September 28, 2022

Active Transfer Prototypical Network: An Efficient Labeling Algorithm for Time-Series Data
Yuqicheng Zhu, Mohamed-Ali Tnani, Timo Jahnz, Klaus Diepold
Active Learning Pre Trained Time Series Data Efficient Learning Prototypical Network Data Labeling Selection Policy

September 23, 2022

From Weakly Supervised Learning to Active Learning
Vivien Cabannes
Active Learning Supervised Learning Weak Supervision Stochastic Gradient Weakly Supervised Learning

September 22, 2022

A Bibliographic View on Constrained Clustering
Ludmila Kuncheva, Francis Williams, Samuel Hennessey
Active Learning Ensemble Learning Automatic Analysis Constrained Clustering Large Scale Comparison Bibliographic Similarity

September 21, 2022

FAL-CUR: Fair Active Learning using Uncertainty and Representativeness on Fair Clustering
Ricky Fajri, Akrati Saxena, Yulong Pei, Mykola Pechenizkiy
Active Learning High Uncertainty Anticipation Fair Clustering Geographical Representativeness Accuracy on the Line Fairness Problem

September 20, 2022

September 14, 2022

Data Lifecycle Management in Evolving Input Distributions for Learning-based Aerospace Applications
Somrita Banerjee, Apoorva Sharma, Edward Schmerling, Max Spolaor, Michael Nemerouf, Marco Pavone
Active Learning Learning Based Model Performance Model Retraining Bayesian Uncertainty Data Management Input Distribution

September 12, 2022

Active Learning and Novel Model Calibration Measurements for Automated Visual Inspection in Manufacturing
Jože M. Rožanec, Luka Bizjak, Elena Trajkova, Patrik Zajec, Jelle Keizer, Blaž Fortuna, Dunja Mladenić
Active Learning Calibration Performance Manufacturing Industry Model Calibration Quality Inspection Calibration Metric Probability Calibration

Active Learning

Papers

VL4Pose: Active Learning Through Out-Of-Distribution Detection For Pose Estimation

Deep Active Ensemble Sampling For Image Classification

Few-Shot Continual Active Learning by a Robot

Is margin all you need? An extensive empirical study of active learning on tabular data

To Softmax, or not to Softmax: that is the question when applying Active Learning for Transformer Models

Making Your First Choice: To Address Cold Start Problem in Vision Active Learning

An Active Learning Reliability Method for Systems with Partially Defined Performance Functions

DIAGNOSE: Avoiding Out-of-distribution Data using Submodular Information Measures

CLINICAL: Targeted Active Learning for Imbalanced Medical Image Classification

Active Learning for Regression with Aggregated Outputs

Nonstationary data stream classification with online active learning and siamese neural networks

Improved Algorithms for Neural Active Learning

Active Transfer Prototypical Network: An Efficient Labeling Algorithm for Time-Series Data

From Weakly Supervised Learning to Active Learning

A Bibliographic View on Constrained Clustering

FAL-CUR: Fair Active Learning using Uncertainty and Representativeness on Fair Clustering

Predictive Scale-Bridging Simulations through Active Learning

Calibrating Ensembles for Scalable Uncertainty Quantification in Deep Learning-based Medical Segmentation

Data Lifecycle Management in Evolving Input Distributions for Learning-based Aerospace Applications

Active Learning and Novel Model Calibration Measurements for Automated Visual Inspection in Manufacturing