BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//132.216.98.100//NONSGML kigkonsult.se iCalcreator 2.20.4//
BEGIN:VEVENT
UID:20260908T132441EDT-0322opvZve@132.216.98.100
DTSTAMP:20260908T172441Z
DESCRIPTION:Informal Systems Seminar (ISS) Centre for Intelligent Machines 
 (CIM) and Groupe d'Etudes et de Recherche en Analyse des Decisions (GERAD)
 \n\nSpeaker: Marco Bonizzato\n	\n	** Note that this is a hybrid event.\n	** N
 ote that this seminar is at University of Montreal.\n\n\n	Zoom Link\n	Meetin
 g ID: 845 1388 1004       \n	Passcode: VISS\n	\n	\n	Abstract: The nervous syst
 em communicates via electrical signals. Electrical neurostimulation is obt
 ained by positioning electrodes in contact with brain\, spinal cord or ner
 ves and delivering stimuli that will modulate neuronal activity. This powe
 rful technique allows causal investigation of neural circuits\, enabling n
 euroscientific discovery. It also constitutes the biophysical foundation o
 f a class of medical interventions. Neurostimulation always requires preci
 se adjustment of several stimulation parameters\, such as the spatial loca
 tion of the stimulus\, the timing\, as well as the frequency of stimulus d
 elivery. Even in the most cutting-edge applications\, stimulation tuning h
 as been almost exclusively handled manually. The lack of algorithmic frame
 works to control and optimize neurostimulation has hindered scientific dis
 covery. Our program is to transform neurostimulation by introducing an adv
 anced autonomous control layer. We use Gaussian Process-based Bayesian Opt
 imization (GPBO) as an algorithmic framework to tailor and personalize neu
 rostimulation to each individual implant. We show that this framework coul
 d be scaled\, via algorithmic novelties\, to unprecedented neurostimulatio
 n steering capacities: 1) from solely stationary to new non-stationary opt
 imization options\, 2) from single target to multi-target optimization\, 3
 ) from simple outputs to sequences of stimuli. This work will equip neuros
 cientists and designers of medical technology with a toolbox of optimizati
 on methods to scale the next generation of medical technologies well beyon
 d the limits of the present constrained control.\n	\n	Bio: Marco Bonizzato i
 s an Automation and Life Sciences Engineer working in implantable brain-co
 mputer interfaces and neuromodulation technology. He has a double expertis
 e in (a) neural prostheses and (b) machine intelligence and optimization. 
 He is an Assistant Professor of Electrical Engineering at Polytechnique Mo
 ntréal\, Adjunct Professor of Neurosciences at Université de Montréal and 
 Associate Academic Member at Mila - Québec AI institute. He is directing t
 he sciNeurotech Lab. The research goal is developing the entire translatio
 nal arc of new neurostimulation therapies\, aiming at restoring sensorimot
 or function after neurotrauma\, from discovery in rodent to application in
  human medical technology\, tailored and personalized to each user by arti
 ficial learning agents.\n	\n	Event Link\n
DTSTART:20240315T143000Z
DTEND:20240315T153000Z
LOCATION:CA\, Room 4488\, André-Aisenstadt Pavilion \, Campus of the Univer
 sity of Montreal
SUMMARY:Autonomous learning agents for intelligent neurostimulation
URL:https://www.mcgill.ca/cim/channels/event/autonomous-learning-agents-int
 elligent-neurostimulation-356114
END:VEVENT
END:VCALENDAR