BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:AI Center Impact Talk - Marisa Ferrara Boston\, Reins AI & Simthet
 ic AI
DTSTART:20251001T140000
DTEND:20251001T150000
DTSTAMP:20260429T185133Z
UID:22216de414cf0ae7c0ca4e703c789d1ffb5e304a7b2a27293dd659c5
CATEGORIES:Conferences - Seminars
DESCRIPTION:Dr. Marisa Ferrara Boston\nThe talk is jointly organized by t
 he EPFL Parallel Systems Architecture Lab (PARSA) and the EPFL AI Center
 .\n\nHosting professor: Prof. Babak Falsafi\n\nFor logistics purposes\, pl
 ease register here (using your EPFL email address): HERE. \n\nThe talk w
 ill be followed by a coffee session. \n\nTitle\nFrom Monitoring to Adapta
 tion: Simulation and Metrics for Reliable AI Agents\n\nAbstract\nMonitorin
 g AI agents can reveal when systems fail\, but it does not tell us how to 
 adapt them. In regulated and data-restricted domains such as finance and h
 ealthcare\, the challenge is even sharper: failures are consequential\, ye
 t the underlying data is often inaccessible for privacy or compliance reas
 ons. This talk introduces simulation and information-theoretic metrics as 
 tools for bridging monitoring and adaptation. By reconstructing failure co
 nditions in simulation\, and using measures such as mutual information and
  entropy\, we can generate synthetic data that supports the next cycle of 
 training and evaluation. I will share examples from audit\, finance\, and 
 healthcare that illustrate how simulation enables adaptation without direc
 t access to sensitive records. The broader aim is to connect foundational 
 theory with applied methods for building reliable AI agents in domains whe
 re the value to society is high but the data is constrained.\n\nBio\nMaris
 a Ferrara Boston is the founder of Reins AI and Simthetic AI\, where she 
 develops evaluation and simulation frameworks for generative AI systems in
  regulated and high-stakes domains. Her work focuses on monitoring\, simul
 ation\, and information-theoretic metrics to build more reliable AI agents
 \, particularly in data-restricted environments such as audit\, finance\, 
 and healthcare. She advises global consulting firms and industry leaders o
 n post-deployment monitoring and adaptation strategies\, helping them navi
 gate the challenges of building AI systems that can evolve safely under re
 gulatory and privacy constraints. She is also a frequent speaker and teach
 er on AI evaluation\, most recently co-leading the IEEE tutorial on Genera
 tive AI Evaluation Essentials.
LOCATION:ELE 117 https://plan.epfl.ch/?room==ELE%20117 https://epfl.zoom.u
 s/j/65297868321?pwd=bI8XGcdAFammW0ag5bv27sM5oiaBt3.1
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR