ICML2022.md

ICML 2022

Year	Title	Author	Publication	Code	Tasks	Notes	Datasets	Notions
2022	Active Multi-Task Representation Learning	Chen et al.	ICML	-	Image Classification	coefficient to select task, CNNs, multi-task learning, None, Hard	MNIST dataset
2022	Achieving Minimax Rates in Pool-Based Batch Active Learning	Gentile et al.	ICML	-	Theory	Hybrid, DNNs, None, pseudo-labels+Human, Hard	-
2022	Active Learning on a Budget: Opposite Strategies Suit High and Low Budgets	Hacohen et al.	ICML	code	Image Classification	Hybrid, FlexMatch, None, Tra, Hard	CIFAR-10/100, ImageNet-100	we claim that the uncer- tainty principle is only suited for the high-budget regime, while the opposite strategy – the selection of the least am- bivalent points – is suitable for the low-budget regime.
2022	SoQal: Selective Oracle Questioning for Consistency Based Active Learning of Cardiac Signals	Kiyasseh et al.	ICML	-	Oracle	Informativeness, BNNs, None, Pseudo+Human, Hard	PhysioNet 2015, PhysioNet 2017, Cardiology
2022	ActiveHedge: Hedge meets Active Learning	Kumar et al.	ICML	-	multiclass prediction	highly-informative, BNNs, None, Tra, Hard	Synthetic
2022	Constants Matter: The Performance Gains of Active Learning	Mussmann and Dasgupta	ICML	-	logistic regression	Uncertainty, BNNs, None, Tra, Hard	Synthetic Data
2022	Discovering Generalizable Spatial Goal Representations via Graph-based Active Reward Learning	Netanyahu et al.	ICML	-	Spatial Goal Representations	reward , Graph-based, Reinforcement learning, Tra, Hard	Human Experiment,
2022	Convergence of Uncertainty Sampling for Active Learning	Raj et al.	ICML	-	binary classification	Uncertainty, BNNs, None, Tra, Hard	Covertype and letter- binary datasets
2022	Efficiently Learning the Topology and Behavior of a Networked Dynamical System Via Active Queries	Rosenkrantz et al.	ICML	code	Synchronous Dynamical Systems	batch and adaptive query, BNNs, None, Tra, Hard	Synthetic, lastfm, gnutella, astroph, facebook, Deezer	We address the problem of infer- ring both the network topology and the behavior of such a system through active queries.
2022	Metric-Fair Active Learning	Shen et al.	ICML	-	Theory	Margin-based, DNNs, None, Tra, Hard	-
2022	N-Penetrate: Active Learning of Neural Collision Handler for Complex 3D Mesh Deformations	Tan et al.	ICML	-	Complex 3D Mesh Deformation	generate AL, N-Penetrate, multi-view capture, Tra, Hard	SCAPE, MIT Swing, MIT Jump, and AMASS- MPIMosh
2022	Cross-Space Active Learning on Graph Convolutional Networks	Tao et al.	ICML	-	plausible explanation about the usefulness of GCNs	Disagreement Coefficients , GCNs, Multi-View Learning, Tra, Hard	-
2022	GALAXY: Graph-based Active Learning at the Extreme	Zhang et al.	ICML	code	Classification	Margin-based, Graph-based, None, Tra, Hard	CIFAR-10, CIFAR_100, SVHN, PATHMNIST