VSAONLINE. SEASON 11. October 6, 20:00GMT. Rich Pang.

3 views

Skip to first unread message

Evgeny Osipov

unread,

Oct 2, 2025, 7:57:55 AMOct 2

to 'Google Groups' via VSACommunity, Parikshit Ram, Daswin De Silva, Fredrick Angelo Galapon, Cynamon, Josh, MCDONALD, NATHAN R CIV USAF AFMC AFRL/RITB, Ibrahim, Mohamed, Ross Gayler, GEETH R DE MEL, Marco Angioli, Peter Bruza, Colyn Seeley, Jesper Olsen, Wanyu Lei, Leonid Mokrushin, Rocco Martino, Trevor Cohen, Dave Bender, Dmitri Rachkovskij, Giacomo Camposampiero, Fatemeh Asgarinejad, Paxon Frady, Nishan Mills

Dear all,

Welcome to the next talk of Season 11 on VSAONLINE. Rich Pang, Neurotaxis, USA will deliver a talk

” A non-Hebbian code for episodic memory”

Date: October 6, 2025

Time: 20:00 GMT

Zoom: https://ltu-se.zoom.us/j/65564790287

WEB: https://bit.ly/vsaonline

Title: A non-Hebbian code for episodic memory.

Abstract: Episodic memories often form after only a single experience, but how this is achieved in the brain is unknown. Current models remain dominated by Hebbian plasticity (“neurons that fire together wire together”) due to its natural ability to bind representations together, support from in vitro experiments, and rich theoretical basis. Modern experiments, however, strongly challenge the role of Hebbian plasticity in vivo. The empirical plasticity rules best matched to one-shot episodic memory timescales have a strikingly different character, for instance depending only on pre-synaptic activity, as in the hippocampal mossy fiber pathway. Yet how these simpler plasticity rules could encode richly structured episodes is unclear. Here we show that by exploiting high-dimensional neural activity with restricted transitions these rules are in fact well very suited for encoding episodes as paths through complex state spaces—such as those underlying a world model. The resulting memory traces, which we term path vectors, simply sum the visited state representations and yet are highly expressive and can be decoded with an odor-tracking algorithm. Through theory and simulation we show that path vectors are a robust alternative to Hebbian traces, support one-shot sequential and associative recall in a variety of scenarios, and suggest a natural biological basis for policy learning. Path vectors also reveal a simple, brain-inspired solution to the classic VSA/HDC binding problem, relying on purely element-wise addition and leveraging prior state spaces and active reconstruction to encode and recall relational information. This work sheds light on how specific plasticity rules observed in the brain can support one-shot episodic memory formation and provides new support for the highly reconstructive nature of recall.