EnergySix-legged robot developed in Penza for pipeline inspectionCNews · 3dModellingEnhanced Imitation Learning of Model Predictive Control through Importance WeightingarXiv — Systems and Control (eess.SY) · 4dModellingLimited knowledge exchange across biological subfields in agent-based modellingarXiv — Physics and Society (physics.soc-ph) · 4dDigital twinReasoner-checked ontology crosswalks for a national digital twinarXiv API — полнотекстовый поиск "digital twin" · 5dDigital twinDigital twin speeds up reinforcement learning for lab robotsarXiv API — полнотекстовый поиск "digital twin" · 5dDigital twinDataset for deciding between evidence gathering and action in control-system incidentsarXiv — Computational Engineering, Finance, and Science (cs.CE) · 5dDigital twinSparse stochastic twins can outperform exact digital twinsarXiv — Computational Engineering, Finance, and Science (cs.CE) · 5dModellingPluralism of rival models in network models of sciencearXiv — Multiagent Systems (cs.MA) · 5dForecastingEstimation and MPC for balancing a wheel-legged robot under unknown disturbancesarXiv — Systems and Control (eess.SY) · 5dDigital twinConsolidating dispersed DSO data into a digital twin of the distribution gridarXiv API — полнотекстовый поиск "digital twin" · 6dDigital twinIntegrating ML and uncertainty quantification for real-time decisions in digital twinsarXiv API — полнотекстовый поиск "digital twin" · 6dDigital twinBiases in end-to-end digital twins built from chained modelsarXiv API — полнотекстовый поиск "digital twin" · 6d
Biases in end-to-end digital twins built from chained models
US & West ·
The work examines the statistical properties of a Monte Carlo scheme that forwards full predictive draws through a chain of separately fitted stage models. Two-stage analytical examples show persistent interval and coverage inflation and, with quadratic propagation, parameter-dependent mean bias; simulations of four- and eight-stage chains confirm pronounced distortions.
Why it matters
It warns that naively assembling a twin from stage models yields biased predictions and points to where correction is needed in end-to-end modelling.
The work defines a formal criterion for valid mappings between digital twin ontologies: coherence and conservativity of the merged ontology. It is applied to the Information Exchange Standard and thirteen ontologies anchored on BOT, Brick and SAREF4BLDG, through matching, human review, certification and reasoning with HermiT.
TwinBridge builds digital twins from 3D Gaussian splatting and physics simulation to supply policy initialization and dense visual rewards for real-world online learning of a laboratory robot. Simulator task events and geometric progress yield rewards for pretraining, distilled into a visual reward model and combined with execution history during human-in-the-loop refinement.
The authors present TE-GvA, a counterfactual decision dataset for fault-driven incident response on the revised Tennessee Eastman process: 9,908 decision points from 1,990 simulated incidents, each with 4–35 candidate actions labelled via branched rollouts. With labels computed under the true hidden cause, gathering evidence is the best choice in only 1.1–1.3% of decisions.