Abstract:
\r\n
\r\nA large fraction of behavior and cognitive neuroscience have been influenced by reinforcement learning theory. However, behavior also happens in the absence of reward. I will explain how concepts of reinforcement learning can be translated to situations and how novelty and surprise can act as internal brain signals, similar to, but distinct from the TD-error signals used in reinforcement learning [1-4]
\r\n
\r\n[1] W. Gerstner, M. Lehmann, V. Liakoni, and J. Brea (2018)
\r\nEligibility traces and plasticity on behavioral time scales: experimental support of NeoHebbian three-factor learning rules.
\r\nFront. Neural Circuits, 12:53 doi: 10.3389/fncir.2018.00053
\r\n
\r\n[2] A. Modirshanechi, J. Brea, and W. Gerstner (2022)
\r\nA taxonomy of surprise definitions
\r\nJ. Mathem. Psychol. 110:102712, DOI: 10.1016/j.jmp.2022.102712
\r\n
\r\n[3] A. Modirshanechi, W.-H. Lin, H.A. Xu, M. Herzog, and W. Gerstner (2025)
\r\nNovelty as a drive of human exploration in complex stochastic environments
\r\nProc. Natl. Acad. Sci. (U.S.A), 122:e2502193122 DOI 10.1073/pnas.2502193122
\r\n
\r\n[4] H.A. Xu, A. Modirshanechi, M.P. Lehmann, W. Gerstner, M.H. Herzog (2021)
\r\nNovelty is not Surprise: Human exploratory and adaptive behavior in sequential decision-making
\r\nPLoS Comput Biol 17: e1009070. doi: https://doi.org/10.1371/journal.pcbi.1009070
\r\n
\r\n[5] V. Liakoni, A. Modirshanechi, W. Gerstner, and J. Brea (2021)
\r\nLearning in Volatile environments with the Bayes Factor Surprise
\r\nNeural Computation 33: 1-72
Join us for the annual EPFL AI Product Management (AIPM) community event on Agentic AI in Business Operations. This half-day event brings together researchers, industry experts, and practitioners to examine how agentic AI is moving from experimentation to practical use in organizations. Through talks, case studies, and discussions, participants will explore how agentic AI systems work, where they create business value, and what security, governance, and organizational questions must be addressed for responsible adoption.
\r\n
\r\nAs part of the annual EPFL AIPM community series, the event is designed to help participants stay informed on recent developments, exchange perspectives with peers, and reflect on practical next steps for their own organizations.
\r\n
\r\nREGISTRATION REQUIRED
\r\n
\r\nEngage with experts from academia and industry to discuss current developments in agentic AI, emerging use cases, and the opportunities and challenges organisations face when adopting autonomous AI systems. Speakers will include:\r\n
Mini-symposium on
\r\nMicrobes in conflict: Phages, Pathogens, Antibiotics and the Human Microbiota
\r\n
\r\nDate: June 4th 2026, 10am- 12pm at EPFL in room AI 1153
\r\nHost: Prof. Melanie Blokesch