Adaptive consolidation of active inference: excitatory and inhibitory mechanisms for organizing feedforward and feedback memory systems in sleep DOI
Don M. Tucker,

Phan Luu,

Karl Friston

и другие.

Cerebral Cortex, Год журнала: 2025, Номер 35(5)

Опубликована: Май 1, 2025

Abstract Cognitive self-organization rests on activity-dependent plasticity to extend the ontogenetic process of neural differentiation and integration cerebral cortex in each act cognition. This account neurocognitive growth can be formulated terms active inference learning. The organism regulates synaptic connectivity as it seeks its goals actively, through excitatory, feedforward expectancies that manifest species-specific affordances. These adaptive are modified reactively, inhibitory feedback error-correction fit predict environmental encounters. Although behavior, concomitant plasticity, occur during waking, architecture requires ongoing consolidation refinement sleep. We propose memory sleep is a continuation neurodevelopmental proceeds kind inversion waking inference: NREM first refines brain’s representations new, unpredicted experiences implementing mechanisms long-term depression both differentiate stabilize new for consolidation. REM then updates organism’s generative world model light this learning excitatory potentiation plasticity. thereby reconsolidates integrates adaptive, predictions, Bayesian priors effective coping.

Язык: Английский

On Bayesian mechanics: a physics of and by beliefs DOI Creative Commons
Maxwell J. D. Ramstead, Dalton A R Sakthivadivel, Conor Heins

и другие.

Interface Focus, Год журнала: 2023, Номер 13(3)

Опубликована: Апрель 14, 2023

The aim of this paper is to introduce a field study that has emerged over the last decade, called Bayesian mechanics. mechanics probabilistic mechanics, comprising tools enable us model systems endowed with particular partition (i.e. into particles), where internal states (or trajectories states) system encode parameters beliefs about external their trajectories). These allow write down mechanical theories for look as if they are estimating posterior probability distributions causes sensory states. This provides formal language modelling constraints, forces, potentials and other quantities determining dynamics such systems, especially entail on space statistical manifold). Here, we will review state art in literature free energy principle, distinguishing between three ways which been applied path-tracking, mode-tracking mode-matching). We go examine duality principle constrained maximum entropy both lie at heart discuss its implications.

Язык: Английский

Процитировано

90
The free energy principle made simpler but not too simple DOI Creative Commons
Karl Friston, Lancelot Da Costa, Noor Sajid

и другие.

Physics Reports, Год журнала: 2023, Номер 1024, С. 1 - 29

Опубликована: Июнь 1, 2023

This paper provides a concise description of the free energy principle, starting from formulation random dynamical systems in terms Langevin equation and ending with Bayesian mechanics that can be read as physics sentience. It rehearses key steps using standard results statistical physics. These entail (i) establishing particular partition states based upon conditional independencies inherit sparsely coupled dynamics, (ii) unpacking implications this inference (iii) describing paths variational principle least action. Teleologically, offers normative account self-organisation optimal design decision-making, sense maximising marginal likelihood or model evidence. In summary, world systems, we end up sentient behaviour interpreted self-evidencing; namely, self-assembly, autopoiesis active inference.

Язык: Английский

Процитировано

90
Generating meaning: active inference and the scope and limits of passive AI DOI Creative Commons
Giovanni Pezzulo, Thomas Parr, Paul Cisek

и другие.

Trends in Cognitive Sciences, Год журнала: 2023, Номер 28(2), С. 97 - 112

Опубликована: Ноя. 15, 2023

Prominent accounts of sentient behavior depict brains as generative models organismic interaction with the world, evincing intriguing similarities current advances in artificial intelligence (AI). However, because they contend control purposive, life-sustaining sensorimotor interactions, living organisms are inextricably anchored to body and world. Unlike passive learned by AI systems, must capture sensory consequences action. This allows embodied agents intervene upon their worlds ways that constantly put best test, thus providing a solid bedrock is – we argue essential development genuine understanding. We review resulting implications consider future directions for AI.

Язык: Английский

Процитировано

41
‘Snakes and ladders’ in paleoanthropology: From cognitive surprise to skillfulness a million years ago DOI Creative Commons
Héctor M. Manrique, Karl Friston, Michael J. Walker

и другие.

Physics of Life Reviews, Год журнала: 2024, Номер 49, С. 40 - 70

Опубликована: Янв. 21, 2024

A paradigmatic account may suffice to explain behavioral evolution in early Homo. We propose a parsimonious that (1) could particular, frequently-encountered, archeological outcome of behavior Homo — namely, the fashioning Paleolithic stone 'handaxe' from biological theoretic perspective informed by free energy principle (FEP); and (2) regards instances as postdictive or retrodictive, circumstantial corroboration. Our proposal considers humankind evolving self-organizing ecosystem at geological time-scale. offer narrative treatment this self-organization terms FEP. Specifically, we indicate how 'cognitive surprises' underwrite an propensity express sporadic unorthodox anomalous behavior. This co-evolutionary has left us legacy artifacts is reminiscent 'snakes ladders' board game appearances, disappearances, reappearances particular traces When detected Early Middle Pleistocene record, anthropologists archeologists often imagine evidence unusual novel ascending rungs figurative phylogenetic 'ladder' if these corresponded progressive cognitive abilities enabled incremental achievements increasingly innovative technical prowess, culminating ascendancy sapiens. The conjecture overlooks plausible likelihood individual who was atypical among her conspecifics have been disregarded community Hominina (for definition see Appendix 1) failed recognize, imagine, articulate potential advantages adopting hitherto Such failure, well diverse fortuitous demographic accidents, would cause exceptional personal be ignored hence unremembered. It disappear pitfall, down 'snake', it were, evolutionary game; thereby causing discontinuity human presents like puzzle. puzzle discomforts some paleoanthropologists trained natural life sciences. They dismiss it, explaining away with such self-justifying conjectures that, maybe, separate paleospecies differentially possessed different abilities, which, supposedly, for presence absence record skill. argue alternative inherits FEP individual's 'active inference' about its surroundings own responses affords prosaic, deflationary, way outcomes skills humankind.

Язык: Английский

Процитировано

11
Principled Limitations on Self-Representation for Generic Physical Systems DOI Creative Commons
Chris Fields,

James F. Glazebrook,

Michael Levin

и другие.

Entropy, Год журнала: 2024, Номер 26(3), С. 194 - 194

Опубликована: Фев. 24, 2024

The ideas of self-observation and self-representation, the concomitant idea self-control, pervade both cognitive life sciences, arising in domains as diverse immunology robotics. Here, we ask a very general way whether, to what extent, these make sense. Using generic model physical interactions, prove theorem several corollaries that severely restrict applicable notions self-observation, self-control. We show, particular, adding observational, representational, or control capabilities meta-level component system cannot, even principle, lead complete representation whole. conclude self-representation can at best be heuristic, self models general, empirically tested by systems implement them.

Язык: Английский

Процитировано

11
Life, its origin, and its distribution: a perspective from the Conway-Kochen Theorem and the Free Energy Principle DOI Creative Commons
Chris Fields, Michael Levin

Communicative & Integrative Biology, Год журнала: 2025, Номер 18(1)

Опубликована: Фев. 17, 2025

We argue here that the Origin of Life (OOL) problem is not just a chemistry but also, and primarily, cognitive science problem. When interpreted through lens Conway-Kochen theorem Free Energy Principle, contemporary physics characterizes all complex dynamical systems persist time as Bayesian agents. If persistent are to some – perhaps only minimal extent cognitive, alive, or living subset systems? no bright line can be drawn, we re-assess, from this perspective, Fermi paradox Drake equation. conclude improving our abilities recognize communicate with diverse intelligences in embodiments, whether based on familiar biochemistry not, will either resolve obviate OOL

Язык: Английский

Процитировано

2
Regulative development as a model for origin of life and artificial life studies DOI
Chris Fields, Michael Levin

Biosystems, Год журнала: 2023, Номер 229, С. 104927 - 104927

Опубликована: Май 19, 2023

Язык: Английский

Процитировано

23
Thoughts and thinkers: On the complementarity between objects and processes DOI
Chris Fields, Michael Levin

Physics of Life Reviews, Год журнала: 2025, Номер 52, С. 256 - 273

Опубликована: Янв. 18, 2025

Язык: Английский

Процитировано

1
Intrinsic Motivation as Constrained Entropy Maximization DOI Creative Commons
Alex Kiefer

Entropy, Год журнала: 2025, Номер 27(4), С. 372 - 372

Опубликована: Март 31, 2025

“Intrinsic motivation” refers to the capacity for intelligent systems be motivated endogenously, i.e., by features of agential architecture itself rather than learned associations between action and reward. This paper views active inference, empowerment, other formal accounts intrinsic motivation as variations on theme constrained maximum entropy providing a general perspective complementary existing frameworks. The connection free energy empowerment noted in previous literature is further explored, it argued that maximum-occupancy approach practice incorporates an implicit model-evidence constraint.

Язык: Английский

Процитировано

1
How do inner screens enable imaginative experience? Applying the free-energy principle directly to the study of conscious experience DOI Creative Commons
Chris Fields, Mahault Albarracin, Karl Friston

и другие.

Neuroscience of Consciousness, Год журнала: 2025, Номер 2025(1)

Опубликована: Янв. 1, 2025

Abstract This paper examines the constraints that free-energy principle (FEP) places on possible model of consciousness, particularly models attentional control and imaginative experiences, including episodic memory planning. We first rehearse classical quantum formulations FEP, focusing their application to multi-component systems, in which only some components interact directly with external environment. In particular, we discuss role internal boundaries have structure Markov blankets, hence function as information channels between components. then show how this formal supports experience, a focus (i) experience can employ spatio-temporal object-recognition reference frames employed ordinary, non-imaginative (ii) be internally generated but still surprising. conclude by discussing implementation, phenomenology, phylogeny implications large state trait variability humans.

Язык: Английский

Процитировано

1