OOIR: Observatory of International Research

Papers

(The H4-Index of Artificial Intelligence is 31. 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 2021-05-01 to 2025-05-01.)

Article	Citations
A computational model of Ostrom's Institutional Analysis and Development framework	387
Iterative voting with partial preferences	352
Ballooning multi-armed bandits	326
Mitigating robust overfitting via self-residual-calibration regularization	274
Online learning in sequential Bayesian persuasion: Handling unknown priors	165
Diagnosis of intermittent faults in Multi-Agent Systems: An SFL approach	154
Knowledge-driven profile dynamics	130
Planning-based knowing how: A unified approach	112
Primarily about primaries	97
Polynomial combined first-order rewritings for linear and guarded existential rules	85
“That's (not) the output I expected!” On the role of end user expectations in creating explanations of AI systems	82
Transfer learning for collaborative recommendation with biased and unbiased data	78
The distortion of distributed metric social choice	70
An efficient algorithm for counting Markov equivalent DAGs	67
Risk-averse autonomous systems: A brief history and recent developments from the perspective of optimal control	64
GoSafeOpt: Scalable safe exploration for global optimization of dynamical systems	62
Epistemic planning: Perspectives on the special issue	61
Manipulation and peer mechanisms: A survey	51
Defense coordination in security games: Equilibrium analysis and mechanism design	48
An α-regret analysis of adversarial bilateral trade	47
Hyper-heuristics for personnel scheduling domains	45
Fair allocation of indivisible goods: Beyond additive valuations	44
Learning MAX-SAT from contextual examples for combinatorial optimisation	43
Accurate parameter estimation for safety-critical systems with unmodeled dynamics	42
(1+1) genetic programming with functionally complete instruction sets can evolve Boolean conjunctions and disjunctions with arbitrarily small error	37

Reward is enough	37
SensorSCAN: Self-supervised learning and deep clustering for fault diagnosis in chemical processes	36
No free lunch theorem for privacy-preserving LLM inference	36
Acceptance in incomplete argumentation frameworks	35
CureGraph: Contrastive multi-modal graph representation learning for urban living circle health profiling and prediction	32
Spectral clustering with robust self-learning constraints	31
Certified reinforcement learning with logic guidance	31
Temporal inductive path neural network for temporal knowledge graph reasoning	31