susanta's blog

This is a reminder that our next monthly webinar is March 20, 3-4pm EDT. The speaker will be Youssef Marzouk from Massachusetts Institute of Technology. 

Speaker: Youssef Marzouk, Massachusetts Institute of Technology 

Title: Sampling and generative modeling using dynamical representations of transport

I am happy to share our paper: “Gradient flow based phase-field modeling using separable neural networks”, published in Computer Methods in Applied Mechanics and Engineering. 

Monthly Webinar. TTA-Uncertainty Quantification and Probabilistic Modeling.

February 13; 3 pm - 4pm EST

Speaker: Alireza Doostan, University of Colorado, Boulder

Title: Uncertainty Quantification and Generative Modeling Using Multi-fidelity Strategies

Link:    https://a

Webinar

January 16; 3pm EST

Speaker: Amanda Howard (Pacific Northwest National Laboratory)

Title: Multifidelity, domain decomposition, and stacking for improving training for physics-informed networks

Dear Colleagues, 

We are organizing a mini-symposium 801 - Recent Trends in Data-Driven and Computational Modeling of Materials Across Scales: From First Principles Calculations to Mesoscale Physics, at USNCCM18, July 20–24, 2025, Chicago. We invite you to submit an abstract at this symposium. The abstract submission deadline is January 15, 2025. Please find the submission portal here:  https://usnccm18.usacm.org/abstract-submission        

Ponkrshnan Thiagarajan, Susanta Ghosh, "Jensen–Shannon divergence based novel loss functions for Bayesian neural networks", Neurocomputing, 2024, 129115, ISSN 0925-2312.  https://doi.org/10.1016/j.neucom.2024.129115  

Highlights

This is a reminder that our next monthly webinar is December 5, 3-4pm EST. The speaker will be Associate Professor Chao Hu from University of Connecticut. We are hoping the format will promote a lively interactive discussion and engage both junior and senior members of our community. Look forward to seeing you there.

Monthly Webinar by USACM, TTA-Uncertainty Quantification and Probabilistic Modeling

December 5; 3pm EST

Speaker: Associate Professor Chao Hu, University of Connecticut

Monthly Webinar: USACM, TTA-Nanotechnology and Lower Scale Phenomena

This is a reminder that our next monthly webinar is November 27, 2-3pm CST. The speaker will be Professor Gábor Csányi from the University of Cambridge. We are hoping the format will promote a lively interactive discussion and engage both junior and senior members of our community. Look forward to seeing you there. Talk and discussions will be archived and available on our website if accepted by the speaker.

November 27; 2pm CST

Speaker: Prof. Gábor Csányi, University of Cambridge

This is a reminder that our next monthly webinar is October 10, 3-4pm ET, is starting soon. 

Monthly Webinar

October 10; 3pm ET

Speaker: Elizabeth Qian, Georgia Tech

Title: Multifidelity linear regression for scientific machine learning from scarce data

Pathrudkar, S., Thiagarajan, P., Agarwal, S. et al. Electronic structure prediction of multi-million atom systems through uncertainty quantification enabled transfer learning. npj Comput Mater 10, 175 (2024). https://doi.org/10.1038/s41524-024-01305-7 

Ponkrshnan Thiagarajan, Trisha Sain, Susanta Ghosh, "Bayesian calibration and uncertainty quantification of a rate-dependent cohesive zone model for polymer interfaces", Engineering Fracture Mechanics,

Volume 309, 2024, 110374, ISSN 0013-7944, https://doi.org/10.1016/j.engfracmech.2024.110374.

Webinar:  Join Zoom Meeting

Our speaker will be Prof. Michele Ceriotti from EPFL.

We look forward to seeing you there.

Date/Time: September 25; 2-3pm CST

Join Zoom Meeting: (https://us06web.zoom.us/j/84541274013?pwd=ZDhuM09BaUlVVlBuREhEV0sweGFlZz09: Meeting ID: 845 4127 4013/Passcode: 799065)

Date/Time: February 28; 2-3pm CST

Speaker: Gideon Simpson, Drexel University

Discussant: Petr Plechac, University of Delaware

Title: Learning dynamics with random Fourier features

USACM Nanotechnology and Lower Scale Phenomena TTA,  webinar is on Jan 31, at 2-3 pm central time.

Speaker: Prof. Surya R. Kalidindi from Georgia Institute of Technology

We have Prof. Raymundo Arróyave of Texas A&M University as the discussant.

We are hoping the format will promote a lively interactive discussion and engage members of our community. We look forward to seeing you there.  Please forward this to anyone who may be interested.

Join Zoom Meeting

Please forward this message to anyone who you feel might be interested in this position. 

We are looking for a self-motivated Ph.D. student to study Machine Learning enabled computational models for applications in Material Modeling and Computational Mechanics, starting August 2024. This project will draw ideas from various subjects including Bayesian Machine Learning, Scientific Machine Learning, Projection-based model reduction, Computational Mechanics, and Medical Imaging. This is a collaborative project between Michigan Tech and University of California, Los Angeles.

Our upcoming webinar is on September 27 at 2pm central.

Date/Time: September 27; 2-3pm CDT

Join via Zoom: https://us06web.zoom.us/j/84541274013?pwd=ZDhuM09BaUlVVlBuREhEV0sweGFlZz09

(Meeting ID: 845 4127 4013/Passcode: 799065)

Speaker: Kaushik Bhattacharya, California Institute of Technology

Discussant: Miguel Bessa, Brown University

Upcoming USACM Nanotechnology webinar on Wednesday, Aug 30, 2:00 pm - 3:00 pm CDT.

Join Zoom Meeting

https://us06web.zoom.us/j/84541274013?pwd=ZDhuM09BaUlVVlBuREhEV0sweGFlZz09

Meeting ID: 845 4127 4013                Passcode: 799065

The Kullback-Leibler (KL) divergence is widely used in state-of-the-art Bayesian Neural Networks (BNNs) to ap- proximate the posterior distribution of weights. However, the KL divergence is unbounded and asymmetric, which may lead to instabilities during optimization or may yield poor generalizations. To overcome these limitations, we examine the Jensen-Shannon (JS) divergence that is bounded, symmetric, and more general. Towards this, we propose two novel loss functions for BNNs.

Funded Ph.D. position in the Department of Mechanical Engineering- Engineering Mechanics at the Michigan Technological University. Starting June/September 2023. 

Kindly forward this message to anyone who you feel might be interested in this position. 

We are looking for a self-motivated Ph.D. student to study Machine Learning enabled computational models for applications in Material Modeling and Computational Mechanics.