Results 141 to 150 of about 9,437,008 (290)

From pixels to planning: scale-free active inference

Frontiers in Network Physiology
This paper describes a discrete state-space model and accompanying methods for generative modeling. This model generalizes partially observed Markov decision processes to include paths as latent variables, rendering it suitable for active inference and ...
Karl Friston, Karl Friston, Conor Heins, Tim Verbelen, Lancelot Da Costa, Lancelot Da Costa, Tommaso Salvatori, Dimitrije Markovic, Dimitrije Markovic, Alexander Tschantz, Magnus Koudahl, Christopher Buckley, Christopher Buckley, Thomas Parr   +13 more
doaj   +1 more source

Automation and Active Learning for the Multi‐Objective Optimization of Antibody Formulations

Advanced Science, EarlyView.
Successful antibody formulation necessitates balancing factors such as thermal stability, colloidal stability, and viscosity across a vast excipient design space. This work integrates robotic liquid handling, high‐throughput biophysical characterization, and multi‐objective Bayesian optimization in an iterative closed‐loop Design‐Build‐Test‐Learn cycle.
D. Christopher Radford, Matthew Tamasi, Elena Di Mare, Adam J. Gormley   +3 more
wiley   +1 more source

Nonparametric Learning Rules from Bandit Experiments: The Eyes have it! [PDF]

How do people learn? We assess, in a distribution-free manner, subjects?learning and choice rules in dynamic two-armed bandit (probabilistic reversal learning) experiments.
Yingyao Hu, Yutaka Kayaba, Matt Shum
core  

PAC Apprenticeship Learning with Bayesian Active Inverse Reinforcement Learning

CoRR
Presented at RLC 2025; published in RLJ ...
Ondrej Bajgar, Dewi S. W. Gould, Jonathon Liu, Alessandro Abate, Konstantinos Gatsis, Michael A. Osborne   +5 more
openaire   +2 more sources

Human‐Guided Bayesian Optimization Enables High‐Throughput Laser Annealing of Mesoporous SiOx Anodes for Lithium‐Ion Batteries

Advanced Science, EarlyView.
An empirical‐aided active learning framework is developed to optimize high‐throughput laser‐induced photothermal annealing of silicon suboxide anodes. By integrating probabilistic machine learning with empirical domain knowledge, this approach achieves optimal electrochemical performance using limited experiments.
Chaeyoung Park, Yeongje Lee, Sunho Jeong, Eungkyu Lee   +3 more
wiley   +1 more source

Learning Bayesian Networks with the bnlearn R Package [PDF]

bnlearn is an R package (R Development Core Team 2010) which includes several algorithms for learning the structure of Bayesian networks with either discrete or continuous variables.
Marco Scutari
core  

A Global Methane Observation System to Reduce Uncertainty for Anthropogenic and Natural Sources and Sinks for Detecting and Attributing Climate Feedbacks

Advanced Science, EarlyView.
Conceptual illustration of a Global Ecosystem Methane Observing System (GEM‐OS) integrating satellites, aircraft, atmospheric networks, and ecosystem measurements to quantify methane emissions from anthropogenic and natural sources. The multi‐scale observing framework improves source attribution, reduces uncertainty in regional methane budgets, and ...
P. Ciais, S. Peng, J. Chang, F. Li, Q. Zhu, K. Yuan, G. Hugelius, H. Li, Y. Cai, F. Chevallier, K. Tibrewal, E. A. Kort, K. Arndt, J. Watts, B. Buma, N. Besic, P. I. Palmer, H. Cadillo‐Quiroz, E. Euskirchen, M. J. Gondwe, A. Hoyt, R. Jackson, S. Malone, D. Monteverde, S. Natali, M. Ramonet, C. Rey‐Sanchez, L. B. Sagang, E. A. G. Schuur, R. Vargas, R. Varner, Z. Zhang, B. Poulter   +32 more
wiley   +1 more source

Revolutionising Agricultural Sustainability: New ‘Furrow Tillage’ can Mitigate Short‐Term Soil‐to‐Atmosphere CO2 Flux and Promote Soil‐Plant‐Microbe Health

Advanced Science, EarlyView.
Furrow tillage resolves the conventional‐vs.‐no‐tillage trade‐off by simultaneously cutting CO2 efflux to 2.0–3.0 g C m−2 d−1 and unlocking high nutrient availability for the rice rhizosphere. This scalable agronomic solution strengthens soil health, enhances plant physiology, reshapes microbial metabolism, and shifts paddy systems toward a net ...
Arnab Majumdar, Munish Kumar Upadhyay, Argha Ghosh, Rakesh Biswas, Ioly Kotta Loizou, Martin Buck, Mark Tibbett, Biswajit Giri, Debojyoti Moulick, Manoj Kumar Jaiswal, Tarit Roychowdhury   +10 more
wiley   +1 more source

Inductive transfer for learning Bayesian networks [PDF]

, 2009
In several domains it is common to have data from different, but closely related problems. For instance, in manufacturing, many products follow the same industrial process but with different conditions; or in industrial diagnosis, where there is ...
LUIS ENRIQUE SUCAR SUCCAR, EDUARDO FRANCISCO MORALES MANZANARES   +1 more
core  

Active learning of model discrepancy with Bayesian experimental design

Computer Methods in Applied Mechanics and Engineering
Digital twins have been actively explored in many engineering applications, such as manufacturing and autonomous systems. However, model discrepancy is ubiquitous in most digital twin models and has significant impacts on the performance of using those models.
Huchen Yang, Chuanqi Chen, Jin-Long Wu
openaire   +2 more sources