On local intrinsic dimensionality of deformation in complex materials

نویسندگان

چکیده

Abstract We propose a new metric called s -LID based on the concept of Local Intrinsic Dimensionality to identify and quantify hierarchies kinematic patterns in heterogeneous media. measures how outlying grain’s motion is relative its nearest neighbors displacement state space. To demonstrate merits over conventional measure strain, we apply it data individual grain motions set deforming granular materials. Several insights into evolution failure are uncovered. First, reveals hierarchy concurrent deformation bands that prevails throughout loading history. These structures vary not only dominance but also spatial scales. Second, nascent stages pre-failure regime, uncovers system-spanning, criss-crossing bands: microbands for small embryonic-shearbands at large , with former being dominant. At opposite extreme, fully formed shearbands dominate microbands. The novel uncovered from contradict common belief causal sequence where subset coalesce and/or grow form shearbands. Instead, suggests sample lead-up governed by complex symbiosis among these different coexisting structures, which amplifies promotes progressive Third, probed this transition microband-dominated regime shearband-dominated systematically suppressing rotations. found particle rotation be an essential enabler regime. When rotations completely suppressed, prevented: coexist parity.

برای دانلود رایگان متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Characterizing Adversarial Subspaces Using Local Intrinsic Dimensionality

Deep Neural Networks (DNNs) have recently been shown to be vulnerable against adversarial examples, which are carefully crafted instances that can mislead DNNs to make errors during prediction. To better understand such attacks, a characterization is needed of the properties of regions (the so-called ‘adversarial subspaces’) in which adversarial examples lie. We tackle this challenge by charact...

متن کامل

Local Intrinsic Dimensionality Based Features for Clustering

One of the fundamental tasks of unsupervised learning is dataset clustering, to partition the input dataset into clusters composed by somehow “similar” objects that “differ” from the objects belonging to other classes. To this end, in this paper we assume that the different clusters are drawn from different, possibly intersecting, geometrical structures represented by manifolds embedded into a ...

متن کامل

Characterizing Adversarial Subspaces Using Local Intrinsic Dimensionality

Deep Neural Networks (DNNs) have recently been shown to be vulnerable against adversarial examples, which are carefully crafted instances that can mislead DNNs to make errors during prediction. To better understand such attacks, a characterization is needed of the properties of regions (the so-called ‘adversarial subspaces’) in which adversarial examples lie. In particular, effective measures a...

متن کامل

Characterizing Adversarial Subspaces Using Local Intrinsic Dimensionality

Deep Neural Networks (DNNs) have recently been shown to be vulnerable against adversarial examples, which are carefully crafted instances that can mislead DNNs to make errors during prediction. To better understand such attacks, the properties of subspaces in the neighborhood of adversarial examples need to be characterized. In particular, effective measures are required to discriminate adversa...

متن کامل

the effect of authentic listening materials on the listening proficiency of efl learners in the intermediate level

having conducted the experiment and analysed the data, the researcher computed the groups mean scores and variances for the test relating to the research question. as the final atep, a t-test was conodonted for the hypothesis. as noted earlier, the significance level was determined at .05 and .01 respectively. the observed t-value was higher than the critical t-value at. 5 and .01 levels. conse...

15 صفحه اول

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

ژورنال

عنوان ژورنال: Scientific Reports

سال: 2021

ISSN: ['2045-2322']

DOI: https://doi.org/10.1038/s41598-021-89328-8