Dynamic planning in hierarchical active inference DOI Creative Commons
Matteo Priorelli, Ivilin Stoianov

Neural Networks, Journal Year: 2025, Volume and Issue: 185, P. 107075 - 107075

Published: Jan. 8, 2025

By dynamic planning, we refer to the ability of human brain infer and impose motor trajectories related cognitive decisions. A recent paradigm, active inference, brings fundamental insights into adaptation biological organisms, constantly striving minimize prediction errors restrict themselves life-compatible states. Over past years, many studies have shown how animal behaviors could be explained in terms inference - either as discrete decision-making or continuous control inspiring innovative solutions robotics artificial intelligence. Still, literature lacks a comprehensive outlook on effectively planning realistic actions changing environments. Setting ourselves goal modeling complex tasks such tool use, delve topic keeping mind two crucial aspects behavior: capacity understand exploit affordances for object manipulation, learn hierarchical interactions between self environment, including other agents. We start from simple unit gradually describe more advanced structures, comparing recently proposed design choices providing basic examples. This study distances itself traditional views centered neural networks reinforcement learning, points toward yet unexplored direction inference: hybrid representations models.

Language: Английский

Generalized Simultaneous Localization and Mapping (G-SLAM) as unification framework for natural and artificial intelligences: towards reverse engineering the hippocampal/entorhinal system and principles of high-level cognition DOI Creative Commons
Adam Safron, Ozan Çatal, Tim Verbelen

et al.

Frontiers in Systems Neuroscience, Journal Year: 2022, Volume and Issue: 16

Published: Sept. 30, 2022

Simultaneous localization and mapping (SLAM) represents a fundamental problem for autonomous embodied systems, which the hippocampal/entorhinal system (H/E-S) has been optimized over course of evolution. We have developed biologically-inspired SLAM architecture based on latent variable generative modeling within Free Energy Principle Active Inference (FEP-AI) framework, affords flexible navigation planning in mobile robots. primarily focused attempting to reverse engineer H/E-S "design" properties, but here we consider ways principles from robotics may help us better understand nervous systems emergent minds. After reviewing LatentSLAM notable features this control architecture, how realize these functional properties not only physical navigation, also with respect high-level cognition understood as generalized simultaneous (G-SLAM). focus loop-closure, graph-relaxation, node duplication particularly impactful architectural features, suggesting computational phenomena contribute understanding cognitive insight (as proto-causal-inference), accommodation integration into existing schemas), assimilation category formation). All operations can similarly be describable terms structure/category learning multiple levels abstraction. However, adopt an ecological rationality perspective, framing functions orchestrating processes both concrete abstract hypothesis spaces. In navigation/search process, adaptive equilibration between involves balancing tradeoffs exploration exploitation; dynamic equilibrium near optimally realized FEP-AI, wherein governed by expected free energy objective naturally balance model simplicity accuracy. With structure learning, such would involve constructing models categories that are neither too inclusive nor exclusive. propose (generalized) represent some most sources variation individuals, modulators functioning potentially illuminate their significances cybernetic parameters. Finally, discuss contributions G-SLAM provide unifying framework its potential realization artificial intelligences.

Language: Английский

Citations

20
Oversampled and undersolved: Depressive rumination from an active inference perspective DOI
Max Berg, Matthias Feldmann, Lukas Kirchner

et al.

Neuroscience & Biobehavioral Reviews, Journal Year: 2022, Volume and Issue: 142, P. 104873 - 104873

Published: Sept. 15, 2022

Language: Английский

Citations

19
Mapping Husserlian Phenomenology onto Active Inference DOI
Mahault Albarracin, Riddhi J. Pitliya, Maxwell J. D. Ramstead

et al.

Communications in computer and information science, Journal Year: 2023, Volume and Issue: unknown, P. 99 - 111

Published: Jan. 1, 2023

Language: Английский

Citations

12
On efficient computation in active inference DOI Creative Commons
Aswin Paul, Noor Sajid, Lancelot Da Costa

et al.

Expert Systems with Applications, Journal Year: 2024, Volume and Issue: 253, P. 124315 - 124315

Published: May 31, 2024

Biological agents demonstrate a remarkable proficiency in calibrating appropriate scales of planning and evaluation when interacting with their environments. It follows logically that any decision-making algorithm aspiring to neurobiological plausibility must mirror these attributes, particularly regarding computational expenditure the intricacy evaluative processes. However, active inference encounters notable challenges simulating apt behaviours within complex These stem chiefly from its substantial demands intricate task defining agent's behaviour preference. We address through two-fold approach. First, we introduce by using Bellman-optimality principle minimise cost function (i.e., expected free energy). Briefly, recursively compute energy actions reverse temporal sequence significantly reduce complexity. Secondly, inspired Z-learning algorithm, propose novel method learn time-constrained agent preferences. face-validate efficacy grid-world simulations precise model learning planning, even under uncertainty. algorithmic advances create new opportunities for various applications—in neuroscience machine learning.

Language: Английский

Citations

4
Dynamic planning in hierarchical active inference DOI Creative Commons
Matteo Priorelli, Ivilin Stoianov

Neural Networks, Journal Year: 2025, Volume and Issue: 185, P. 107075 - 107075

Published: Jan. 8, 2025

By dynamic planning, we refer to the ability of human brain infer and impose motor trajectories related cognitive decisions. A recent paradigm, active inference, brings fundamental insights into adaptation biological organisms, constantly striving minimize prediction errors restrict themselves life-compatible states. Over past years, many studies have shown how animal behaviors could be explained in terms inference - either as discrete decision-making or continuous control inspiring innovative solutions robotics artificial intelligence. Still, literature lacks a comprehensive outlook on effectively planning realistic actions changing environments. Setting ourselves goal modeling complex tasks such tool use, delve topic keeping mind two crucial aspects behavior: capacity understand exploit affordances for object manipulation, learn hierarchical interactions between self environment, including other agents. We start from simple unit gradually describe more advanced structures, comparing recently proposed design choices providing basic examples. This study distances itself traditional views centered neural networks reinforcement learning, points toward yet unexplored direction inference: hybrid representations models.

Language: Английский

Citations

0