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 2022-01-01 to 2026-01-01.)

Article	Citations
Learning to bid and rank together in recommendation systems	224
The role of mutual information in variational classifiers	180
Robust reputation independence in ranking systems for multiple sensitive attributes	152
Maximum causal entropy inverse constrained reinforcement learning	92
Parameter identifiability of a deep feedforward ReLU neural network	89
Surrogate models of radiative transfer codes for atmospheric trace gas retrievals from satellite observations	83
Adapting performance metrics for ordinal classification to interval scale: length matters	74
Meta-learning for heterogeneous treatment effect estimation with closed-form solvers	70
Compositional scene modeling with global object-centric representations	66
Modelling spatiotemporal dynamics from Earth observation data with neural differential equations	63
Fairness seen as global sensitivity analysis	46
SETAR-Tree: a novel and accurate tree algorithm for global time series forecasting	41
Automated imbalanced classification via layered learning	41
Resolving power: a general approach to compare the distinguishing ability of threshold-free evaluation metrics	40
Spike2CGR: an efficient method for spike sequence classification using chaos game representation	39
Semantic-enhanced graph neural networks with global context representation	36
Correction to: Segmentation and feature extraction-based classification of pavement damages using hybrid computer vision and machine learning approaches	35
On the usefulness of the fit-on-test view on evaluating calibration of classifiers	34
Transfer learning with pre-trained conditional generative models	33
Differentially-private data synthetisation for efficient re-identification risk control	33
Simultaneous outlier detection and elimination in hyperspectral unmixing via weighted non-negative matrix tri-factorization	33
GENs: generative encoding networks	32
Masked autoencoder for multiagent trajectories	32
Towards accurate knowledge transfer via target-awareness representation disentanglement	32
Efficient private SCO for heavy-tailed data via averaged clipping	31

Trimming stability selection increases variable selection robustness	31
Responsible model deployment via model-agnostic uncertainty learning	28
A calibration test for evaluating set-based epistemic uncertainty representations	28
LaMMOn: language model combined graph neural network for multi-target multi-camera tracking in online scenarios	28