BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Joint MechE Colloquium & Civil Engineering Seminar: Meta-neural To
 pology Optimization: knowledge infusion in Engineering design
DTSTART:20250513T120000
DTEND:20250513T130000
DTSTAMP:20261010T024325Z
UID:0679c398bc8f0a073e130532d2b2a085959df3d60b2110144fc15766
CATEGORIES:Conferences - Seminars
DESCRIPTION:Prof. Miguel Bessa\, Brown School of Engineering\, Brown Univ
 ersity\nAbstract: Engineers learn from every design they create\, buildin
 g intuition that helps them quickly identify promising solutions for new p
 roblems. Unfortunately\, topology optimization methods for Engineering des
 ign ignore past knowledge: every problem starts from a “blank canvas”.
  We propose a new method\, meta-neural topology optimization\, that infuse
 s knowledge from past examples via a meta-learning strategy. The trained n
 eural network creates effective designs that evolve towards optimal soluti
 ons faster than existing methods. Importantly\, the proposed method does n
 ot suffer from “hallucination” issues that plague generative machine l
 earning models\, which would lead to catastrophic consequences in Engineer
 ing. Every design it creates is predicted with a simulation obeying Physic
 s principles and leads to consistent designs\, akin to conventional topolo
 gy optimization methods. We believe that this fusion of machine learning a
 nd topology optimization will be a key enabler of future Engineering disco
 veries.\n\nBiography: Professor Miguel Bessa and his research group envis
 ion a new era for the design of materials and structures using artificial 
 intelligence. Professor Bessa received a PhD in Mechanical Engineering fro
 m Northwestern University in 2016 as a Fulbright scholar. After a short po
 stdoctoral position at Caltech (2017) and a quick leap from Assistant to A
 ssociate Professor (2021) at Delft University of Technology\, he joined th
 e Solid Mechanics Group at Brown University in the Summer of 2022.
LOCATION:MED 0 1418 https://plan.epfl.ch/?room==MED%200%201418 https://epf
 l.zoom.us/j/64267570786
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR