Batch Active Learning

Batch active learning aims to efficiently train machine learning models by strategically selecting batches of unlabeled data points for labeling, maximizing model performance with minimal annotation effort. Current research focuses on developing novel algorithms that balance informativeness and diversity within these batches, employing techniques like determinantal point processes, generative flow networks, and Bayesian methods, often tailored to specific model architectures (e.g., Gaussian processes, neural networks). This approach is particularly valuable in resource-constrained scenarios, improving the efficiency of model training across diverse applications such as speech recognition, robotics, and medical image analysis. The resulting advancements enhance data utilization and reduce the cost associated with data labeling.

Papers

August 3, 2024

Batch Active Learning in Gaussian Process Regression using Derivatives
Hon Sum Alec Yu, Christoph Zimmer, Duy Nguyen-Tuong
Gaussian Process Gaussian Process Regression Total Correlation Derivative Process Batch Active Learning

May 3, 2024

Combining X-Vectors and Bayesian Batch Active Learning: Two-Stage Active Learning Pipeline for Speech Recognition
Ognjen Kundacina, Vladimir Vincan, Dragisa Miskovic
Automatic Speech Recognition Active Learning Speech Recognition Speech Data X Vector Batch Active Learning Learning Pipeline

February 24, 2024

Batch Active Learning of Reward Functions from Human Preferences
Erdem Bıyık, Nima Anari, Dorsa Sadigh
Reward Function Robot Learning Human Preference Active Learning Algorithm Batch Active Learning

July 19, 2023

Novel Batch Active Learning Approach and Its Application to Synthetic Aperture Radar Datasets
James Chapman, Bohan Chen, Zheng Tan, Jeff Calder, Kevin Miller, Andrea L. Bertozzi
Application Proficiency Active Learning Synthetic Aperture Radar Sequential Learning Batch Active Learning

June 26, 2023

BatchGFN: Generative Flow Networks for Batch Active Learning
Shreshth A. Malik, Salem Lahlou, Andrew Jesson, Moksh Jain, Nikolay Malkin, Tristan Deleu, Yoshua Bengio, Yarin Gal
Active Learning Generative Flow Network Pool Based Active Learning Batch Active Learning

May 10, 2023

Accelerating Batch Active Learning Using Continual Learning Techniques
Arnav Das, Gantavya Bhatt, Megh Bhalerao, Vianne Gao, Rui Yang, Jeff Bilmes
Active Learning Catastrophic Forgetting Continual Learning Replay Based Continual Learning Batch Active Learning Continual Learning Paradigm

February 17, 2023

Black-Box Batch Active Learning for Regression
Andreas Kirsch
Active Learning Novel Regression Black Box Model Regression Task Batch Active Learning

December 3, 2022

Active learning using adaptable task-based prioritisation
Shaheer U. Saeed, João Ramalhinho, Mark Pinnock, Ziyi Shen, Yunguan Fu, Nina Montaña-Brown, Ester Bonmati, Dean C. Barratt, Stephen P. Pereira, Brian Davidson, Matthew J. Clarkson, Yipeng Hu
Active Learning Segmentation Task Segmentation Network Vessel Segmentation Priority Based Batch Active Learning

November 1, 2022

Batch Active Learning from the Perspective of Sparse Approximation
Maohao Shen, Bowen Jiang, Jacky Yibo Zhang, Oluwasanmi Koyejo
Active Learning Visual Perspective Model Training Sparsity Constraint Sparse Approximation Batch Active Learning

June 20, 2022

Deep reinforced active learning for multi-class image classification
Emma Slade, Kim M. Branson
Active Learning Medical Image Classification Active Learning Framework Batch Active Learning

March 17, 2022

A Framework and Benchmark for Deep Batch Active Learning for Regression
David Holzmüller, Viktor Zaverkin, Johannes Kästner, Ingo Steinwart
New Benchmark New Framework Active Learning Gaussian Process Supervised Learning Novel Regression Neural Tangent Kernel Deep Regression Batch Active Learning

December 27, 2021

Scalable Batch-Mode Deep Bayesian Active Learning via Equivalence Class Annealing
Renyu Zhang, Aly A. Khan, Robert L. Grossman, Yuxin Chen
Active Learning Deep Active Learning Latent Class Mini Batch Consistency ActIve Learning algORithm Batch Active Learning Decision Aware