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

September 20, 2023

Large-scale Pretraining Improves Sample Efficiency of Active Learning based Molecule Virtual Screening
Zhonglin Cao, Simone Sciabola, Ye Wang
Active Learning Drug Discovery Sample Efficiency Virtual Screening Large Scale Pretraining Compound Library Ligand Based Based Virtual Screening

September 18, 2023

September 17, 2023

Active Learning for Semantic Segmentation with Multi-class Label Query
Sehyun Hwang, Sohyun Lee, Hoyoung Kim, Minhyeon Oh, Jungseul Ok, Suha Kwak
Semantic Segmentation Active Learning Segmentation Model Partial Label Learning Multi Class

September 16, 2023

September 15, 2023

Active Learning for Fine-Grained Sketch-Based Image Retrieval
Himanshu Thakur, Soumitri Chattopadhyay
Fine Grained Active Learning Sketch Based Image Retrieval Sketch Photo

September 14, 2023

September 12, 2023

Annotating Data for Fine-Tuning a Neural Ranker? Current Active Learning Strategies are not Better than Random Selection
Sophia Althammer, Guido Zuccon, Sebastian Hofstätter, Suzan Verberne, Allan Hanbury
Pre Trained Language Model Active Learning Pretrained Language Model Data Annotation Neural Ranker Annotation Budget Random Selection

September 11, 2023

September 8, 2023

Active Learning for Classifying 2D Grid-Based Level Completability
Mahsa Bazzaz, Seth Cooper
Active Learning

September 5, 2023

Optimal Sample Selection Through Uncertainty Estimation and Its Application in Deep Learning
Yong Lin, Chen Liu, Chenlu Ye, Qing Lian, Yuan Yao, Tong Zhang
Deep Learning Application Proficiency Active Learning Uncertainty Estimation Benchmark Datasets Sample Allocation

September 4, 2023

On the Query Strategies for Efficient Online Active Distillation
Michele Boldo, Enrico Martini, Mirco De Marchi, Stefano Aldegheri, Nicola Bombieri
Deep Learning Knowledge Distillation Active Learning Human Pose Estimation Real Time Adaptation Online Distillation Query Strategy

September 2, 2023

Streaming Active Learning for Regression Problems Using Regression via Classification
Shota Horiguchi, Kota Dohi, Yohei Kawaguchi
Classification Code Active Learning Novel Regression Regression Problem Regression via Classification

September 1, 2023

Explainable Active Learning for Preference Elicitation
Furkan Cantürk, Reyhan Aydoğan
Active Learning Preference Elicitation Active Learning With Contrastive Explanation Limited Labeled Data

August 28, 2023

Active Learning

Papers

Large-scale Pretraining Improves Sample Efficiency of Active Learning based Molecule Virtual Screening

Efficient Concept Drift Handling for Batch Android Malware Detection Models

Active anomaly detection based on deep one-class classification

Active Learning for Semantic Segmentation with Multi-class Label Query

Interactively Teaching an Inverse Reinforcement Learner with Limited Feedback

Enhancing personalised thermal comfort models with Active Learning for improved HVAC controls

Active Learning for Fine-Grained Sketch-Based Image Retrieval

ALWOD: Active Learning for Weakly-Supervised Object Detection

Semantic Parsing in Limited Resource Conditions

Annotating Data for Fine-Tuning a Neural Ranker? Current Active Learning Strategies are not Better than Random Selection

ChemSpaceAL: An Efficient Active Learning Methodology Applied to Protein-Specific Molecular Generation

Stream-based Active Learning by Exploiting Temporal Properties in Perception with Temporal Predicted Loss

Learning Objective-Specific Active Learning Strategies with Attentive Neural Processes

Active Learning for Classifying 2D Grid-Based Level Completability

Optimal Sample Selection Through Uncertainty Estimation and Its Application in Deep Learning

On the Query Strategies for Efficient Online Active Distillation

Streaming Active Learning for Regression Problems Using Regression via Classification

Explainable Active Learning for Preference Elicitation

Maturity-Aware Active Learning for Semantic Segmentation with Hierarchically-Adaptive Sample Assessment

Distributionally Robust Statistical Verification with Imprecise Neural Networks