سخنران: F. Ghanbarnejad, Sharif University of Technology

8 تیر 1401
18:00 تا 19:30

Here I will review my recent works [1-14] on modeling interacting contagious dynamics, for example coupled SIR or SIS dynamics, in mean field approximations and also on different random generated or empirical complex networks. I show and discuss how our recent results have been improving our understanding and prediction of epidemic dynamics and disease ecology while raising new questions and challenges in physics of critical phenomena.

Also I will briefly discuss SARS-COV-2 from the perspective of disease ecology and present my recent studies on "behavioral responses to the COVID-19 spread"; which focus on the analysis and modelling of empirical data including testing strategies, air traffic, urban mobility, inequality (socio-economic classes), news spread, etc. in Iran or USA or globally.

سخنران: هادی حشمی پور ، پژوهشگاه دانش های بنیادی

8 تیر 1401
11:00 تا 12:00

آنالیز شبکه عصبی تابع ترکش هادرون باردار سبک در حضور داده های
SIDIS

سخنران: احمد شیخی
دانشگاه شیراز

8 تیر 1401
13:30 تا 15:00

این سمینار روز چهارشنبه مورخ 8 تیر ماه ساعت 13:30 به صورت آنلاین برگزار می شود.

https://www.skyroom.online/ch/soa/weekly-seminar

سخنران: دکتر علی صادقی

8 تیر 1401
13:00 تا 14:00

چکیده:

A class of supervised machine learning approaches aims at predicting a quantity from an input data vector. For example, it is common practice to recognize a person's face from the set of data points (pixels) in a digital image frame. The same techniques are effectively useful in computational condensed matter physics problems for the prediction of atomic contributions to a given physical quantity from the arrangement of the neighboring atoms of the individual atoms. Then one needs a "descriptor" that quantifies the environment of an atom such that it can be fed as input to a supervised machine learning tool. We review the basic ideas, techniques, and challenges of fingerprinting the local environment as an essential ingredient of training interatomic potentials from e.g. ab initio reference samples. We exemplify the local fingerprinting by applying it to predict the excited and ground-state energy surfaces of large atomic systems.


برای شرکت در این وبینار وارد لینک زیر شده و به صورت میهمان لاگین کنید


https://www.skyroom.online/ch/schoolofnanoscience/weeklyseminars

سخنران: A.Sadeghi, Physics Department, Shahid Beheshti University

8 تیر 1401
13:00 تا 14:00

Date: Wednesday, June 29, 2022 /چهارشنبه ، ۸ تیر ۱۴۰۱
Time: 13:00-14:00

To join the webinar, please follow this link and log in as a guest:
https://www.skyroom.online/ch/schoolofnanoscience/weeklyseminars

Abstract:

A class of supervised machine learning approaches aims at predicting a quantity from an input data vector. For example, it is common practice to recognize a person's face from the set of data points (pixels) in a digital image frame. The same techniques are effectively useful in computational condensed matter physics problems for the prediction of atomic contributions to a given physical quantity from the arrangement of the neighboring atoms of the individual atoms. Then one needs a "descriptor" that quantifies the environment of an atom such that it can be fed as input to a supervised machine learning tool. We review the basic ideas, techniques, and challenges of fingerprinting the local environment as an essential ingredient of training interatomic potentials from e.g. ab initio reference samples. We exemplify the local fingerprinting by applying it to predict the excited and ground-state energy surfaces of large atomic systems.

سخنران: ندجا والنتین, موسسه ریاضیات دوسلدورف آلمان

8 تیر 1401
15:00 تا 17:00

In a joint work with Chernikov we have shown that the structure of non-degenerate bilinear forms on an NIP field are 2-dependent. One of the main ingredients is a composition lemma, which issues that NIP formulas remain 2-dependent when precomposed with binary functions. In this talk, we will discuss possible generalizations of non-degeneracy for n-linear forms and the decomposition lemma, present the structure of non-degenerate n-linear forms and draw the connection to n-dependent structures.
(joint work with Chernikov)

https://us06web.zoom.us/j/9086116889?pwd=WGRFOGZWZ1FOMXJrcWpJMWFqUFIvQT09
Meeting ID: 908 611 6889
Passcode: 362880

سخنران: Sh.Abdolrahimi, California State Polytechnic University

7 تیر 1401
14:00 تا 15:00


Date and time

Tuesday, 28th of June 2022 (7th of Tir, 1401), 2:00 pm (Tehran zone)


Link
https://us02web.zoom.us/j/83538126663?pwd=VTNQWVpOdDZ1ajAvbVI1WklKODZGUT09



Meeting ID: 835 3812 6663
Passcode: 785277


Abstract
In this talk, I will analyze the concept of a distorted black hole.
As examples, I focus on the case of distorted black holes in higher dimensions and the most common generating techniques. In particular, I represent the distorted black ring solution and some of its interesting properties.

سخنران: یاسمن حسینی، پژوهشگاه دانش های بنیادی

6 تیر 1401
11:00 تا 12:00

جستجوی فوتون های تاریک در تولید بوزون هیگز از طریق همجوشی بوزون برداری در برخورد پروتون-پروتون در
$\sqrt{s}=13\TeV$

سخنران: فاطمه باقری فرد, پژوهشگاه دانشهای بنیادی

1 تیر 1401
14:00 تا 15:00

Scientific computing is a broad, multidisciplinary area that encompasses mathematics, computer science, and applications in science and engineering. This area makes use of the techniques of applied mathematics and computer science for the solution of scientific and engineering problems which are modeled by some differential equations. More recently, solving differential equations via learning methods emerged as a potentially new sub-field under the name of Scientific Machine Learning (SciML). This work advocates a fruitful synergy between machine learning and classical computational physics that has the potential to enrich both fields and lead to high-impact developments. In this talk, we review some learning methods such as support vector machines and neural networks that lead to some numerical learning approaches. The support vector machine model for classification tasks can be modified for the regression problems. Neural networks as well as deep networks can also be used to efficiently approximate a function, derived from the universal approximation theorems.

To get more information about the Combinatorics and Computing Weekly Seminar and webinars Zoom link, join the following google group:
https://groups.google.com/g/ipmccs

سخنران: مهدی هرمزی, پژوهشگاه دانشهای بنیادی

1 تیر 1401
16:00 تا 17:00

Singular integral operators are among important tools in harmonic analysis. In this talk, I will discuss the history and origins of these operators and give some examples. I will also describe the so-called ``Weight theory'' with an overview of some recent results on this subject.

To get more information about the colloquium, join the following google group: https://groups.google.com/g/ipm-math-colloquium

سخنران: واسکو براتکا, دانشگاه مونیخ، آلمان

1 تیر 1401
15:30 تا 17:30

We introduce the discontinuity problem and we prove that under the axiom of determinacy it is actually the weakest discontinuous problem in the topological Weihrauch lattice. We also discuss algebraic properties of the discontinuity problem, such as the fact that it parallelizes to the non-computability problem. We also introduce an operation that we call stashing and show that is dual to parallelization. We show that the discontinuity problem is the stashing of LLPO and several variants of it.

Zoom info:
https://us02web.zoom.us/j/83113646825?pwd=NVpsdERQSXkxWk9McXc1eFFISFc4QT09
Meeting ID: 831 1364 6825
Passcode: 365815

سخنران: جلیل صداقت، دانشگاه شیراز

1 تیر 1401
11:00 تا 12:00

ساختار ستاره ی کوارکی شگفت مغناطیده در رهیافت کرومودینامیک اختلالی