OOIR: Observatory of International Research

Papers

(The H4-Index of Machine Learning is 28. The table below lists those papers that are above that threshold based on CrossRef citation counts [max. 250 papers]. The publications cover those that have been published in the past four years, i.e., from 2021-11-01 to 2025-11-01.)

Article	Citations
Automated imbalanced classification via layered learning	188
One-Stage Tree: end-to-end tree builder and pruner	144
Learning to bid and rank together in recommendation systems	121
The role of mutual information in variational classifiers	83
Robust reputation independence in ranking systems for multiple sensitive attributes	72
Meta-learning for heterogeneous treatment effect estimation with closed-form solvers	72
Maximum causal entropy inverse constrained reinforcement learning	72
Parameter identifiability of a deep feedforward ReLU neural network	65
Resolving power: a general approach to compare the distinguishing ability of threshold-free evaluation metrics	60
A review on instance ranking problems in statistical learning	59
Spike2CGR: an efficient method for spike sequence classification using chaos game representation	59
Surrogate models of radiative transfer codes for atmospheric trace gas retrievals from satellite observations	39
Adapting performance metrics for ordinal classification to interval scale: length matters	39
Fairness seen as global sensitivity analysis	38
Semantic-enhanced graph neural networks with global context representation	37
Compositional scene modeling with global object-centric representations	35
Modelling spatiotemporal dynamics from Earth observation data with neural differential equations	32
SETAR-Tree: a novel and accurate tree algorithm for global time series forecasting	31
A prompt-driven framework for multi-domain knowledge tracing	31
On the usefulness of the fit-on-test view on evaluating calibration of classifiers	31
An accelerated proximal algorithm for regularized nonconvex and nonsmooth bi-level optimization	30
FairSwiRL: fair semi-supervised classification with representation learning	30
Maintaining AUC and H-measure over time	30
A flexible class of dependence-aware multi-label loss functions	29
GENs: generative encoding networks	29

Simultaneous outlier detection and elimination in hyperspectral unmixing via weighted non-negative matrix tri-factorization	29
Transfer learning with pre-trained conditional generative models	29
Differentially-private data synthetisation for efficient re-identification risk control	29