Filters








58,670 Hits in 5.9 sec

Potential Energy and Particle Interaction Approach for Learning in Adaptive Systems [chapter]

Deniz Erdogmus, Jose C. Principe, Luis Vielva, David Luengo
2002 Lecture Notes in Computer Science  
In this paper, inspired by this idea, we propose a generalization to the particle interaction model for learning and system adaptation.  ...  Recently, Principe et al. proposed a particle interaction model for information theoretical learning.  ...  In this paper, inspired by the idea of information theoretic learning through particle interactions, we have proposed an alternative approach to adaptation and learning.  ... 
doi:10.1007/3-540-46084-5_74 fatcat:4drnippj55bgncfuoizjs2w3hm

Inverse methods for design of soft materials [article]

Zachary M. Sherman, Michael P. Howard, Beth A. Lindquist, Ryan B. Jadrich, Thomas M. Truskett
2020 arXiv   pre-print
In this Perspective, we discuss recent advances in inverse methods for design of soft materials that address two challenges: (1) methodological limitations that prevent such approaches from satisfying  ...  design constraints and (2) computational challenges that limit the size and complexity of systems that can be addressed.  ...  Because µ cannot be controlled in real systems, µ was set to zero to avoid biasing the particle shape, and the external tethering potential was slowly reduced.  ... 
arXiv:2004.00181v1 fatcat:djvudcwdnrfh3kiofq4pxxbafy

Adaptive Machine Learning for Time-Varying Systems: Towards 6D Phase Space Diagnostics of Short Intense Charged Particle Beams [article]

Alexander Scheinker, Spencer Gessner
2022 arXiv   pre-print
New adaptive machine learning (AML) methods designed for time-varying systems are needed to aid in the diagnostics and control of high-intensity, ultrashort beams by combining deep learning tools such  ...  Although machine learning (ML) methods have grown in popularity in the accelerator community in recently years, they are fundamentally limited when it comes to time-varying systems for which most current  ...  A broad overview of recent developments and the state of the art in adaptive controls and machine learning for particle accelerators can be found in the proceedings of the 2019 Advanced Control Methods  ... 
arXiv:2203.04391v2 fatcat:sxmcvh4kprfpnps6q2wom7qfem

Molecular Learning of a Soft-Disks Fluid [article]

Luca Zammataro
2021 bioRxiv   pre-print
I describe the fluid's learning as the property of an order that emerges as an adaptation in establishing equilibrium with energy and thermal conservation.  ...  It provides learning proofs in a Lennard-Jones (LJ) fluid, presented as a network of particles having non-bonded interactions.  ...  In these last, kinetic and potential energy concepts interpret learning as a dissipation-driven adaptation mechanism.  ... 
doi:10.1101/2021.07.24.453642 fatcat:yacp7jphmff7rmkqqtn346qoky

Optimising Particle Accelerators with Adaptive Machine Learning

2020 Research Outreach  
Dr Scheinker works on an adaptive machine learning approach that has the potential to optimise any complex time-varying system.  ...  varies unpredictably over time. of adaptive feedback for stabilisation optimisation, control of unknown timevarying systems, and adaptive machine learning for time-varying systems such as particle accelerator  ... 
doi:10.32907/ro-115-182185 fatcat:ybheeozxpbhbnhymr74ha6qn2a

Self-Learning Adaptive Umbrella Sampling Method for the Determination of Free Energy Landscapes in Multiple Dimensions

Wojciech Wojtas-Niziurski, Yilin Meng, Benoı̂t Roux, Simon Bernèche
2013 Journal of Chemical Theory and Computation  
potential defined as a Fermat spiral and to a model system consisted of Lennard-Jones particles.  ...  in accuracy. * Corresponding authors: simon.berneche@unibas.ch, roux@uchicago.edu. † Those two authors contributed equally to this work Application of the self-learning adaptive US approach to an analytical  ...  Acknowledgments S.B. is grateful to Guillaume Lamoureux for fruitful discussions in the early stage of development of the method.  ... 
doi:10.1021/ct300978b pmid:23814508 pmcid:PMC3694627 fatcat:rzsbjx6aefcofptwbuc5lkjvfe

Thermodynamics of Evolution and the Origin of Life [article]

Vitaly Vanchurin, Yuri I. Wolf, Eugene V. Koonin, Mikhail I. Katsnelson
2021 arXiv   pre-print
We then develop a phenomenological approach to the description of evolution, which involves modeling the grand potential as a function of the biological temperature and evolutionary potential.  ...  (macroscopic counterpart of fitness) and free energy (macroscopic counterpart of additive fitness).  ...  In Sec. 4 we develop a phenomenological approach to evolution and define relevant thermodynamic potentials (such as average loss function, free energy, grand potential) and thermodynamic parameters (such  ... 
arXiv:2110.15066v1 fatcat:bfn6n5xnmbfozdv3ftznkfhvna

Chemically specific coarse‐graining of polymers: Methods and prospects

Satyen Dhamankar, Michael A. Webb
2021 Journal of Polymer Science  
In this review, we discuss essential modeling techniques, bottom-up coarse-graining methodologies, and outstanding challenges for the chemically specific CG modeling of polymer-based systems.  ...  Given continuing advancements in experimental synthesis and characterization of such systems, there is ever greater need to leverage and expand CG capabilities to simulate diverse soft matter systems with  ...  Gartner III, and anonymous reviewers for critical readings of the manuscript.  ... 
doi:10.1002/pol.20210555 fatcat:3v54qbozazeuzdf2p5h33vp2hq

On the Autotuning Potential of Time-stepping methods from Scientific Computing

Natalia Kalinnik, Robert Kiesel, Thomas Rauber, Marcel Richter, Gudula Rünger
2018 Proceedings of the 2018 Federated Conference on Computer Science and Information Systems  
In this article, we explore the autotuning potential of several methods from scientific computing.  ...  We also address the question, whether offline or online autotuning approaches are appropriate for the specific method.  ...  Acknowledgement This work has been supported by the German Ministry of Science and Education (BMBF), Project title SeASiTe (Self-Adaptation of Time-step-based Simulation Techniques on Heterogeneous HPC  ... 
doi:10.15439/2018f169 dblp:conf/fedcsis/KalinnikKRRR18 fatcat:ld2hg3qk7vhdhbdfxmlhrpd73e

Guest Editorial for the Special Section on Brain Computer Interface (BCI)

Dongrui Wu, Brent J. Lance, Vernon J. Lawhern
2017 IEEE transactions on fuzzy systems  
The breadth of the research captured by these articles provides an indication of the importance of BCI in human-machine interaction and indicates the potential for further development of fuzzy logic in  ...  The adaptive, self-organizing and interpretable modeling framework based on general type-2 fuzzy sets is able to learn in real-time without prior training.  ...  TRANSACTIONS ON FUZZY SYSTEMS, for his suggestions and advice throughout the entire process of this special issue.  ... 
doi:10.1109/tfuzz.2017.2652799 fatcat:hqxvoefwdbc2hl4ioiss6tqc6q

A molecular dynamics based digital twin for ultrafast laser material removal processes

Panagiotis Stavropoulos, Alexios Papacharalampopoulos, Lydia Athanasopoulou
2020 The International Journal of Advanced Manufacturing Technology  
Within the latest years, digital twins have become one of the most promising concepts that can be applied to complex manufacturing processes, due to their accuracy and adaptiveness in real-time what-if  ...  The simulation responses are integrated into a digital twin utilizing machine learning techniques, physics and decision-making algorithms.  ...  The determination of the inter-atomic potential and the potential energy in the particles of the systems is being obtained by the Morse potential function (MPF) [14] .  ... 
doi:10.1007/s00170-020-05387-7 fatcat:p5hdeacinjf4dmmgqgbf3j4jd4

Neuroevolutionary Learning of Particles and Protocols for Self-Assembly

Stephen Whitelam, Isaac Tamblyn
2021 Physical Review Letters  
The learning algorithm is capable of both directed and exploratory design: it can assemble a material with a user-defined property, or search for novelty in the space of specified order parameters.  ...  Within simulations of molecules deposited on a surface we show that neuroevolutionary learning can design particles and time-dependent protocols to promote self-assembly, without input from physical concepts  ...  In pursuit of "synthesis by design" [1, 2] the materials science community has developed and adapted algorithms of inverse design and machine learning.  ... 
doi:10.1103/physrevlett.127.018003 fatcat:askqygkte5bpzfydaknoixnjry

Illustrating chaos: a schematic discretization of the general three-body problem in Newtonian gravity

Nathan W C Leigh, Shalma Wegsman
2018 Monthly notices of the Royal Astronomical Society  
Each time a transformation is applied, the system changes state as the particles re-distribute their energy and angular momenta.  ...  transformation in energy- and angular momentum-space, thereby potentially mitigating any computational expense.  ...  We further thank Johan Samsing and Nicholas Stone for an early read of our manuscript and critical feedback. N. W. C.  ... 
doi:10.1093/mnras/sty192 fatcat:ad66hzqt75a55aazkplr65peve

Learning DFT [article]

Peter Schmitteckert
2020 arXiv   pre-print
Instead of applying machine learning to the energy functional itself, we apply these techniques to the Kohn-Sham potentials.  ...  Finally, we use the neural network to up-scale the simulation to larger system sizes.  ...  ACKNOWLEDGMENTS Most of the work reported here was performed while being at the university of Würzburg and was supported by ERC-StG-Thomale-TOPOLECTRICS-336012 and was presented at the FQMT'19 in Praque  ... 
arXiv:2008.07923v1 fatcat:4tlupzktgrbrnkakaxyo4wdy3u

Modern computational studies of the glass transition [article]

Ludovic Berthier, David R. Reichman
2022 arXiv   pre-print
We finally describe some important challenges for future research.  ...  Supercooled liquids and glasses have been studied numerically since the advent of molecular dynamics and Monte Carlo simulations in the last century.  ...  The authors thank all of the members of the Simons Foundation Collaboration on "Cracking the Glass Problem" for six+ years of stimulating discussions.  ... 
arXiv:2208.02206v1 fatcat:7nvqilopebbmjevryf2ezg2xdq
« Previous Showing results 1 — 15 out of 58,670 results