| 2022 |
Active Surrogate Estimators: An Active Learning Approach to Label-Efficient Model Evaluation |
Kossen et al. |
NIPS |
code |
Active Surrogate Estimators |
|
|
|
| 2022 |
Batch Multi-Fidelity Active Learning with Budget Constraints |
Li et al. |
NIPS |
- |
Computational physics and Engineering applications |
Diversity, Bayesian NN, None, Tra, Hard |
physical simulation (solving Poisson’s, Heat and viscous Burger’s equations), a topology structure design problem, and a computational fluid dynamics (CFD) task to predict the velocity field of boundary-driven flows. |
|
| 2022 |
Active Learning for Multiple Target Models |
Tang and Huang |
NIPS |
- |
OCR |
disagreement-based, 12 specialized model architectures, None, Pre-FT,Hard |
MNIST, Kuzushiji-MNIST |
|
| 2022 |
Few-Shot Continual Active Learning by a Robot |
Ayub and Fendley |
NIPS |
- |
object classification |
Uncertainty, Gaussian mixture model , continue learning, |
CORe-50 |
|
| 2022 |
Making Look-Ahead Active Learning Strategies Feasible with Neural Tangent Kernels |
Mohamadi et al. |
NIPS |
- |
Image classification |
Influence, DNNs, None, Tra, Hard |
MNIST, SVHN, CIFAR10, CIFAR100 |
|
| 2022 |
Active Learning of Classifiers with Label and Seed Queries |
Bressan et al. |
NIPS |
- |
Theory |
|
|
|
| 2022 |
Improved Algorithms for Neural Active Learning |
Ban et al. |
NIPS |
- |
non-parametric streaming setting |
|
|
|
| 2022 |
Active Learning Through a Covering Lens |
Yehuda et al. |
NIPS |
code |
Image classification |
Representative, CNNs, None, FT, Hard |
CIFAR-10, CIFAR-100, Tiny-ImageNet, ImageNet |
|
| 2022 |
Active Learning Helps Pretrained Models Learn the Intended Task |
Tamkin et al. |
NIPS |
- |
Pre-traing+AL |
Uncertainty, BiT+Roberta, None, Pre-FT,Hard |
|
|
Waterbirds, Treeperson, iWildCam2020-WILDS, Amazon-WILDS |
|
|
|
|
|
|
|
|
| 2022 |
Deep Active Learning by Leveraging Training Dynamics |
Wang et al. |
NIPS |
- |
Image Classification |
Train-speed CNN,REsNet,VGG, None, Pre-FT, Hard |
CIFAR10, SVHN, Caltech101 |
|
| 2022 |
A Lagrangian Duality Approach to Active Learning |
Elenter et al. |
NIPS |
- |
classification, Regression |
Informativeness, ResNet-18,Constrained learning,Pre-FT,Hard |
STL-10 [54], CIFAR-10 [55], SVHN [56] and MNIST |
|
| 2022 |
Active Learning Polynomial Threshold Functions |
Ben-Eliezer et al. |
NIPS |
- |
Theory |
Derivative queries, |
|
improve lower bound of AL |
| 2022 |
Active Learning with Safety Constraints |
Camilleri et al. |
NIPS |
- |
best-arm identification in linear bandits with safety constraints |
Baysian, Any, None, Tra, Hard |
German Credit dataset, Half circle dataset |
find the best arm satisfying certain (unknown) safety constraints |
| 2022 |
Active Learning with Neural Networks: Insights from Nonparametric Statistics |
Zhu and Nowak |
NIPS |
- |
Theory |
|
|
minimax label complexity |
| 2022 |
Efficient Active Learning with Abstention |
Zhu and Nowak |
NIPS |
- |
Theory |
|
|
break the computational barrier and design an efficient active learning algorithm |
| 2022 |
Meta-Query-Net: Resolving Purity-Informativeness Dilemma in Open-set Active Learning |
Park et al. |
NIPS |
- |
Image Classification |
Informativeness, MLP,Meta-Learning,Tra,Hard |
CIFAR10, CIFAR100, ImageNet |
filtering out the noisy examples |