A k‐means clustering machine learning‐based multiscale method for anelastic heterogeneous structures with internal variables

2022 ◽  
Author(s):  
M. A. Benaimeche ◽  
J. Yvonnet ◽  
B. Bary ◽  
Q.‐C. He
Keyword(s):  
Machine Learning ◽  
Multiscale Method ◽  
Internal Variables ◽  
Heterogeneous Structures

Thermodynamic-informed machine learning for pressure-sensitive plasticity model

2021 ◽  
Author(s):  
WaiChing Sun ◽  
Nikolas Vlassis
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Plastic Flow ◽  
Yield Function ◽  
Supervised Machine Learning ◽  
Internal Variables ◽  
Elasto Plasticity ◽  
Path Dependent ◽  
Speed Function ◽  
Low Dimensional

<p>This talk will present a machine learning framework that builds interpretable macroscopic surrogate elasto-plasticity models inferred from sub-scale direction numerical simulations (DNS) or experiments with limited data. To circumvent the lack of interpretability of the classical black-box neural network, we introduce a higher-order supervised machine learning technique that generates components of elasto-plastic models such as elasticity functional, yield function, hardening mechanisms, and plastic flow. The geometrical interpretation in the principal stress space allows us to use convexity and smoothness to ensure thermodynamic consistency. The speed function from the Hamilton-Jacobi equation is deduced from the DNS data to formulate hardening and non-associative plastic flow rules governed by the evolution of the low-dimensional descriptors. By incorporating a non-cooperative game that determines the necessary data to calibrate material models, the machine learning generated model is continuously tested, calibrated, and improved as new data guided by the adversarial agents are generated. A graph convolutional neural network is used to deduce low-dimensional descriptors that encodes the evolutional of particle topology under path-dependent deformation and are used to replace internal variables. The resultant constitutive laws can be used in a finite element solver or incorporated as a loss function for the physical-informed neural network run physical simulations.</p>

A machine-learning-enhanced hierarchical multiscale method for bridging from molecular dynamics to continua

2019 ◽  
Vol 32 (18) ◽  
pp. 14359-14373 ◽  
Author(s):  
Shaoping Xiao ◽  
Renjie Hu ◽  
Zhen Li ◽  
Siamak Attarian ◽  
Kaj-Mikael Björk ◽  
...  
Keyword(s):  
Machine Learning ◽  
Molecular Dynamics ◽  
Multiscale Method ◽  
Hierarchical Multiscale

Mind wandering as data augmentation: How mental travel supports abstraction

2020 ◽  
Vol 43 ◽  
Author(s):  
Myrthe Faber
Keyword(s):  
Machine Learning ◽  
Data Augmentation ◽  
Mental Content ◽  
Mind Wandering ◽  
Theoretical Framework ◽  
Important Addition

Abstract Gilead et al. state that abstraction supports mental travel, and that mental travel critically relies on abstraction. I propose an important addition to this theoretical framework, namely that mental travel might also support abstraction. Specifically, I argue that spontaneous mental travel (mind wandering), much like data augmentation in machine learning, provides variability in mental content and context necessary for abstraction.

Membrane–cytoskeleton interactions in cholesterol-dependent domain formation

2015 ◽  
Vol 57 ◽  
pp. 177-187 ◽  
Author(s):  
Jennifer N. Byrum ◽  
William Rodgers
Keyword(s):  
Biological Properties ◽  
Membrane Rafts ◽  
Membrane Domains ◽  
Biological Functions ◽  
Membrane Cytoskeleton ◽  
Heterogeneous Structures ◽  
Integral Role ◽  
Model Cell ◽  
Actomyosin Cytoskeleton ◽  
Dependent Domain

Since the inception of the fluid mosaic model, cell membranes have come to be recognized as heterogeneous structures composed of discrete protein and lipid domains of various dimensions and biological functions. The structural and biological properties of membrane domains are represented by CDM (cholesterol-dependent membrane) domains, frequently referred to as membrane ‘rafts’. Biological functions attributed to CDMs include signal transduction. In T-cells, CDMs function in the regulation of the Src family kinase Lck (p56lck) by sequestering Lck from its activator CD45. Despite evidence of discrete CDM domains with specific functions, the mechanism by which they form and are maintained within a fluid and dynamic lipid bilayer is not completely understood. In the present chapter, we discuss recent advances showing that the actomyosin cytoskeleton has an integral role in the formation of CDM domains. Using Lck as a model, we also discuss recent findings regarding cytoskeleton-dependent CDM domain functions in protein regulation.

Sign in / Sign up

Export Citation Format

Share Document