Active Inference Research Articles

Kinetic Energy and the Free Energy Principle in the Birth of Human Life

Kinetic Energy and the Free Energy Principle in the Birth of Human Life

Form, function, mind: What doesn't compute (and what might)

The applicability of computational and dynamical systems models to organisms is scrutinized, using examples from developmental biology and cognition. Developmental morphogenesis is dependent on the inherent material properties of developing animal (metazoan) tissues, a non-computational modality, but cell differentiation, which utilizes chromatin-based revisable memory banks and program-like function-calling, via the developmental gene co-expression system unique to the metazoans, has a quasi-computational basis. Multi-attractor dynamical models are argued to be misapplied to global properties of development, and it is suggested that along with computationalism, classic forms of dynamicism are similarly unsuitable to accounting for cognitive phenomena. Proposals are made for treating brains and other nervous tissues as novel forms of excitable matter with inherent properties which enable the intensification of cell-based basal cognition capabilities present throughout the tree of life. Finally, some connections are drawn between the viewpoint described here and active inference models of cognition, such as the Free Energy Principle.

The free energy principle induces intracellular compartmentalization

Living systems at all scales are compartmentalized into interacting subsystems. This paper reviews a mechanism that drives compartmentalization in generic systems at any scale. It first discusses three symmetries of generic physical interactions in a quantum-theoretic description. It then shows that if one of these, a permutation symmetry on the inter-system boundary, is spontaneously broken, the symmetry breaking is amplified by the Free Energy Principle (FEP). It thus shows how compartmentalization generically results from permutation symmetry breaking under the FEP. It finally notes that the FEP asymptotically restores the broken symmetry, showing that the FEP can be regarded as a theory of fluctuations away from a permutation-symmetric boundary, and hence from an entangled joint state of the interacting systems.

Framing the predictive mind: why we should think again about Dreyfus

AbstractIn this paper I return to Hubert Dreyfus’ old but influential critique of artificial intelligence, redirecting it towards contemporary predictive processing models of the mind (PP). I focus on Dreyfus’ arguments about the “frame problem” for artificial cognitive systems, and his contrasting account of embodied human skills and expertise. The frame problem presents as a prima facie problem for practical work in AI and robotics, but also for computational views of the mind in general, including for PP. Indeed, some of the issues it presents seem more acute for PP, insofar as it seeks to unify all cognition and intelligence, and aims to do so without admitting any cognitive processes or mechanisms outside of the scope of the theory. I contend, however, that there is an unresolved problem for PP concerning whether it can both explain all cognition and intelligent behavior as minimizing prediction error with just the core formal elements of the PP toolbox, and also adequately comprehend (or explain away) some of the apparent cognitive differences between biological and prediction-based artificial intelligence, notably in regard to establishing relevance and flexible context-switching, precisely the features of interest to Dreyfus’ work on embodied indexicality, habits/skills, and abductive inference. I address several influential philosophical versions of PP, including the work of Jakob Hohwy and Andy Clark, as well as more enactive-oriented interpretations of active inference coming from a broadly Fristonian perspective.

Sustainability under Active Inference

In this paper, we explore the known connection among sustainability, resilience, and well-being within the framework of active inference. Initially, we revisit how the notions of well-being and resilience intersect within active inference before defining sustainability. We adopt a holistic concept of sustainability denoting the enduring capacity to meet needs over time without depleting crucial resources. It extends beyond material wealth to encompass community networks, labor, and knowledge. Using the free energy principle, we can emphasize the role of fostering resource renewal, harmonious system–entity exchanges, and practices that encourage self-organization and resilience as pathways to achieving sustainability both as an agent and as a part of a collective. We start by connecting active inference with well-being, building on existing work. We then attempt to link resilience with sustainability, asserting that resilience alone is insufficient for sustainable outcomes. While crucial for absorbing shocks and stresses, resilience must be intrinsically linked with sustainability to ensure that adaptive capacities do not merely perpetuate existing vulnerabilities. Rather, it should facilitate transformative processes that address the root causes of unsustainability. Sustainability, therefore, must manifest across extended timescales and all system strata, from individual components to the broader system, to uphold ecological integrity, economic stability, and social well-being. We explain how sustainability manifests at the level of an agent and then at the level of collectives and systems. To model and quantify the interdependencies between resources and their impact on overall system sustainability, we introduce the application of network theory and dynamical systems theory. We emphasize the optimization of precision or learning rates through the active inference framework, advocating for an approach that fosters the elastic and plastic resilience necessary for long-term sustainability and abundance.

Natural language syntax complies with the free-energy principle

Natural language syntax yields an unbounded array of hierarchically structured expressions. We claim that these are used in the service of active inference in accord with the free-energy principle (FEP). While conceptual advances alongside modelling and simulation work have attempted to connect speech segmentation and linguistic communication with the FEP, we extend this program to the underlying computations responsible for generating syntactic objects. We argue that recently proposed principles of economy in language design—such as “minimal search” criteria from theoretical syntax—adhere to the FEP. This affords a greater degree of explanatory power to the FEP—with respect to higher language functions—and offers linguistics a grounding in first principles with respect to computability. While we mostly focus on building new principled conceptual relations between syntax and the FEP, we also show through a sample of preliminary examples how both tree-geometric depth and a Kolmogorov complexity estimate (recruiting a Lempel–Ziv compression algorithm) can be used to accurately predict legal operations on syntactic workspaces, directly in line with formulations of variational free energy minimization. This is used to motivate a general principle of language design that we term Turing–Chomsky Compression (TCC). We use TCC to align concerns of linguists with the normative account of self-organization furnished by the FEP, by marshalling evidence from theoretical linguistics and psycholinguistics to ground core principles of efficient syntactic computation within active inference.

Abstract PO2-15-11: Kinetic Analysis of [5-11C] glutamine PET tracer data in breast cancer: Preclinical studies in mouse models

Abstract Background: Glutamine is an important metabolic substrate in many aggressive tumors, with comparable importance to glucose metabolism. Utilizing human breast cancer mouse xenograft models, we studied the kinetics of the PET imaging agent, [5-11C] glutamine, a biochemical authentic substrate for glutamine metabolism, to further characterize the metabolism of glutamine and downstream labeled metabolites. Studies were performed with and without inhibition of the enzyme, glutaminase (GLS), the first step in glutamine catabolism that generates glutamate, and key target for therapy directed to glutamine-metabolizing cancers. Methods: The study used xenograft mouse models for two breast cancer cell lines, HCC1806, a highly glutaminolytic triple-negative cell line, and MCF-7, a hormone receptor positive line with only low levels of glutaminolysis. Mice were injected with 5-[11C]L-glutamine, and the contributions of individual metabolites to the total [11C] signal in blood and tumor tissue were estimated at 10, 20, and 30 minutes after injection by assaying the fractional contributions of [5-11C] glutamine and labeled metabolites, specifically [5-11C] glutamate, [11C] CO2 and all other metabolites with [11C], in both blood and tumor tissue. This generated estimated time activity curves for [5-11C] glutamine and downstream metabolites in both HCC1806 and MCF-7 mouse models with and without CB-839 treatment, an inhibitor of GLS. We also investigated a compartment model that describes these data, including model simulations that fit the model to match the measured curves. Results: We found that, out to 30 minutes post-injection, the vast majority of tumor radioactivity was in the form of either 5-11C] glutamine or [5-11C] glutamate, with smaller amounts of radioactivity in the form of more downstream metabolites, including [11C]CO2. Systemic administration of the glutaminase inhibitor alters the arterial input curve by reducing the systemic rate of conversion of [5-11C] glutamine to [5-11C] glutamate. However, in GLS active tumors, [5-11C] glutamate retained in the large cellular glutamate pool represents a greater fraction of total radioactivity than in the blood and leads to a total tumor time-activity curve that is only marginally different than the total tumor time-activity curve after GLS inhibition. Data from blood and tissue analysis were used to test a hypothetical kinetic model for [5-11C] glutamine, which provided estimates of glutamate and glutamate tumor tissue concentrations and tumor GLS activity that agree with direct measurements in the animal models. Conclusion: The study of [ 5-11C] glutamine in breast cancer models leads to kinetic insight that are in line with biochemical studies and direct measurement taken for tumor tissue obtained from the animal models. Retention of metabolized [5-11C] glutamine as [5-11C] glutamate in a large tissue pool in tumors with high GLS activity makes non-invasive inference of GLS activity challenging without additional ex vivo analysis, supporting the use of non-metabolized glutamine analogs such as such as [18F](2S,4R)4-fluoroglutamine ([18F]4F-Gln) to infer in vivo glutamine metabolism. Citation Format: Christopher Hensley, Prashanth Padakanti, Hoon Choi, Christina Dulal, Hsiaojiu Lee, Austin Pantel, Rong Zhou, David Mankoff. Kinetic Analysis of [5-11C] glutamine PET tracer data in breast cancer: Preclinical studies in mouse models [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO2-15-11.

Active Inference With Empathy Mechanism for Socially Behaved Artificial Agents in Diverse Situations.

This article proposes a method for an artificial agent to behave in a social manner. Although defining proper social behavior is difficult because it differs from situation to situation, the agent following the proposed method adaptively behaves appropriately in each situation by empathizing with the surrounding others. The proposed method is achieved by incorporating empathy into active inference. We evaluated the proposed method regarding control of autonomous mobile robots in diverse situations. From the evaluation results, an agent controlled by the proposed method could behave more adaptively socially than an agent controlled by the standard active inference in the diverse situations. In the case of two agents, the agent controlled with the proposed method behaved in a social way that reduced the other agent's travel distance by 13.7% and increased the margin between the agents by 25.8%, even though it increased the agent's travel distance by 8.2%. Also, the agent controlled with the proposed method behaved more socially when it was surrounded by altruistic others but less socially when it was surrounded by selfish others.

Imagination vs. routines: festive time, weekly time, and the predictive brain.

This paper examines the relationship between societal structures shaped by traditions, norms, laws, and customs, and creative expressions in arts and media through the lens of the predictive coding framework in cognitive science. The article proposes that both dimensions of culture can be viewed as adaptations designed to enhance and train the brain's predictive abilities in the social domain. Traditions, norms, laws, and customs foster shared predictions and expectations among individuals, thereby reducing uncertainty in social environments. On the other hand, arts and media expose us to simulated experiences that explore alternative social realities, allowing the predictive machinery of the brain to hone its skills through exposure to a wider array of potentially relevant social circumstances and scenarios. We first review key principles of predictive coding and active inference, and then explore the rationale of cultural traditions and artistic culture in this perspective. Finally, we draw parallels between institutionalized normative habits that stabilize social worlds and creative and imaginative acts that temporarily subvert established conventions to inject variability.

An active inference approach to interpersonal differences in depression

Depression is characterized by different distortions in interpersonal experience and behavior, ranging from social withdrawal to overt hostility. However, clinical psychological research has largely neglected the need for an integrative framework to operationalize these different phenomena and their dynamic change more accurately in depression. In this article, we draw on active inference theory, a comprehensive theory of perception, action, and learning, to provide a formal model explaining how variations in patients' internal belief-systems lead to differences in social experience and behavior. In this context, we assume that individuals cannot directly grasp the characteristics of their social environment. Instead, they must infer them indirectly from ambiguous social observations, which they themselves generate and alter through their actions. Differences in interpersonal experience and behavior arise from the interplay of patients’ prior expectations, their propensity to infer particular social states from certain observations, and their beliefs in their ability to influence these situations through specific actions. We then use concrete examples to demonstrate how future research can take our approach to identify systematic differences in interpersonal experiences and behaviors among depressed patients (or patient subgroups) and to investigate their changes in response to new social experiences. We also discuss potential applications of our approach in diagnosing and treating depression. This work is a move towards understanding the interpersonal aspects of depression in more detail, recognizing their importance in etiology, diagnosis, and treatment.

Crafting explainable artificial intelligence through active inference: A model for transparent introspection and decision-making

This paper explores the feasibility of constructing interpretable artificial intelligence (AI) systems rooted in active inference and the free energy principle. Initially, we offer a concise introduction to active inference, emphasizing its relevance to modeling decision-making, introspection, and the generation of both overt and covert actions. Subsequently, we delve into how active inference can serve as a foundation for designing explainable AI systems. Specifically, it enables us to capture essential aspects of "introspective" processes and generate intelligible models of decision-making mechanisms. We propose an architectural framework for explainable AI systems employing active inference. Central to this framework is an explicit hierarchical generative model that enables the AI system to monitor and elucidate the factors influencing its decisions. Importantly, this model's structure is designed to be understandable and verifiable by human users. We elucidate how this architecture can amalgamate diverse data sources to make informed decisions in a transparent manner, mirroring aspects of human consciousness and introspection. Finally, we examine the implications of our findings for future AI research and discuss potential ethical considerations associated with developing AI systems with (apparent) introspective capabilities.

Crafting explainable artificial intelligence through active inference: A model for transparent introspection and decision-making

This paper explores the feasibility of constructing interpretable artificial intelligence (AI) systems rooted in active inference and the free energy principle. Initially, we offer a concise introduction to active inference, emphasizing its relevance to modeling decision-making, introspection, and the generation of both overt and covert actions. Subsequently, we delve into how active inference can serve as a foundation for designing explainable AI systems. Specifically, it enables us to capture essential aspects of "introspective" processes and generate intelligible models of decision-making mechanisms. We propose an architectural framework for explainable AI systems employing active inference. Central to this framework is an explicit hierarchical generative model that enables the AI system to monitor and elucidate the factors influencing its decisions. Importantly, this model's structure is designed to be understandable and verifiable by human users. We elucidate how this architecture can amalgamate diverse data sources to make informed decisions in a transparent manner, mirroring aspects of human consciousness and introspection. Finally, we examine the implications of our findings for future AI research and discuss potential ethical considerations associated with developing AI systems with (apparent) introspective capabilities.

An Overview of the Free Energy Principle and Related Research.

The free energy principle and its corollary, the active inference framework, serve as theoretical foundations in the domain of neuroscience, explaining the genesis of intelligent behavior. This principle states that the processes of perception, learning, and decision making-within an agent-are all driven by the objective of "minimizing free energy," evincing the following behaviors: learning and employing a generative model of the environment to interpret observations, thereby achieving perception, and selecting actions to maintain a stable preferred state and minimize the uncertainty about the environment, thereby achieving decision making. This fundamental principle can be used to explain how the brain processes perceptual information, learns about the environment, and selects actions. Two pivotal tenets are that the agent employs a generative model for perception and planning and that interaction with the world (and other agents) enhances the performance of the generative model and augments perception. With the evolution of control theory and deep learning tools, agents based on the FEP have been instantiated in various ways across different domains, guiding the design of a multitude of generative models and decision-making algorithms. This letter first introduces the basic concepts of the FEP, followed by its historical development and connections with other theories of intelligence, and then delves into the specific application of the FEP to perception and decision making, encompassing both low-dimensional simple situations and high-dimensional complex situations. It compares the FEP with model-based reinforcement learning to show that the FEP provides a better objective function. We illustrate this using numerical studies of Dreamer3 by adding expected information gain into the standard objective function. In a complementary fashion, existing reinforcement learning, and deep learning algorithms can also help implement the FEP-based agents. Finally, we discuss the various capabilities that agents need to possess in complex environments and state that the FEP can aid agents in acquiring these capabilities.

Collective behavior from surprise minimization

Collective motion is ubiquitous in nature; groups of animals, such as fish, birds, and ungulates appear to move as a whole, exhibiting a rich behavioral repertoire that ranges from directed movement to milling to disordered swarming. Typically, such macroscopic patterns arise from decentralized, local interactions among constituent components (e.g., individual fish in a school). Preeminent models of this process describe individuals as self-propelled particles, subject to self-generated motion and "social forces" such as short-range repulsion and long-range attraction or alignment. However, organisms are not particles; they are probabilistic decision-makers. Here, we introduce an approach to modeling collective behavior based on active inference. This cognitive framework casts behavior as the consequence of a single imperative: to minimize surprise. We demonstrate that many empirically observed collective phenomena, including cohesion, milling, and directed motion, emerge naturally when considering behavior as driven by active Bayesian inference-without explicitly building behavioral rules or goals into individual agents. Furthermore, we show that active inference can recover and generalize the classical notion of social forces as agents attempt to suppress prediction errors that conflict with their expectations. By exploring the parameter space of the belief-based model, we reveal nontrivial relationships between the individual beliefs and group properties like polarization and the tendency to visit different collective states. We also explore how individual beliefs about uncertainty determine collective decision-making accuracy. Finally, we show how agents can update their generative model over time, resulting in groups that are collectively more sensitive to external fluctuations and encode information more robustly.

Privacy preservation in sensor-based Human Activity Recognition through autoencoders for low-power IoT devices

Human activity recognition is increasingly recognized as a key task in many applications. However, gathering data from the variety of sensors commonly available on end devices risks compromising user’s privacy when signals are transmitted to more powerful computing units for inference offloading. It is therefore important to design and implement strategies that could prevent privacy breaches without impairing the capability of the system of recognizing activity patterns, and by taking into account the energy constraints of low-power devices. In this work, we propose an energy-aware approach aimed at preserving the privacy of users during inference of human activities. The proposed method is based on a deep learning autoencoder trained to process the signal in order to remove the most sensitive information regarding privacy attributes, without significantly impacting classification accuracy. We also perform a thorough architecture’s parameter tuning of the designed system to enable its implementation on a low-power platform, which we also characterize in terms of energy expenditure. Experimental results show that this system is capable of effectively transforming the signal in order to prevent the inference of sensitive attributes (i.e. weight, height, age, and gender) and it can be conveniently implemented on a constrained embedded system at different levels of the trade-off between accuracy and energy consumption. Indeed, a complete obfuscation of sensitive attributes can be achieved at the cost of a marginal reduction in classification accuracy (5% at most), with an expenditure of around 165 mJ for an execution time of around 30ms needed during the signal transformation step.

Feeling our place in the world: an active inference account of self-esteem.

Self-esteem, the evaluation of one's own worth or value, is a critical aspect of psychological well-being and mental health. In this paper, we propose an active inference account of self-esteem, casting it as a sociometer or an inferential capacity to interpret one's standing within a social group. This approach allows us to explore the interaction between an individual's self-perception and the expectations of their social environment.When there is a mismatch between these perceptions and expectations, the individual needs to adjust their actions or update their self-perception to better align with their current experiences. We also consider this hypothesis in relation with recent research on affective inference, suggesting that self-esteem enables the individual to track and respond to this discrepancy through affective states such as anxiety or positive affect. By acting as an inferential sociometer, self-esteem allows individuals to navigate and adapt to their social environment, ultimately impacting their psychological well-being and mental health.

Modeling Motor Control in Continuous Time Active Inference: A Survey

Modeling Motor Control in Continuous Time Active Inference: A Survey

Shared Protentions in Multi-Agent Active Inference.

In this paper, we unite concepts from Husserlian phenomenology, the active inference framework in theoretical biology, and category theory in mathematics to develop a comprehensive framework for understanding social action premised on shared goals. We begin with an overview of Husserlian phenomenology, focusing on aspects of inner time-consciousness, namely, retention, primal impression, and protention. We then review active inference as a formal approach to modeling agent behavior based on variational (approximate Bayesian) inference. Expanding upon Husserl's model of time consciousness, we consider collective goal-directed behavior, emphasizing shared protentions among agents and their connection to the shared generative models of active inference. This integrated framework aims to formalize shared goals in terms of shared protentions, and thereby shed light on the emergence of group intentionality. Building on this foundation, we incorporate mathematical tools from category theory, in particular, sheaf and topos theory, to furnish a mathematical image of individual and group interactions within a stochastic environment. Specifically, we employ morphisms between polynomial representations of individual agent models, allowing predictions not only of their own behaviors but also those of other agents and environmental responses. Sheaf and topos theory facilitates the construction of coherent agent worldviews and provides a way of representing consensus or shared understanding. We explore the emergence of shared protentions, bridging the phenomenology of temporal structure, multi-agent active inference systems, and category theory. Shared protentions are highlighted as pivotal for coordination and achieving common objectives. We conclude by acknowledging the intricacies stemming from stochastic systems and uncertainties in realizing shared goals.

Joint Distortion Restoration and Quality Feature Learning for No-reference Image Quality Assessment

No-reference image quality assessment (NR-IQA) methods, inspired by the free energy principle, improve the accuracy of image quality prediction by simulating the human brain’s repair process for distorted images. However, existing methods use separate optimization schemes for distortion restoration and quality prediction, which undermines the accurate mapping of feature representations to quality scores. To address this issue, we propose a joint restoration and quality feature learning NR-IQA (RQFL-IQA) method to jointly tackle distortion image restoration and quality prediction within a unified framework. To accurately establish the quality reconstruction relationship between distorted and restored images, a hybrid loss function based on pixel-wise and structure-wise representations is used to improve the restoration capability of the image restoration network. The proposed RQFL-IQA exploits rich labels, including restored images and quality scores, to enable the model to learn more discriminative features and establish a more accurate mapping from feature representation to quality scores. In addition, to avoid the impact of poor restoration on quality prediction, we propose a module with a cleaning function to reweight the fusion of restored and primitive features to achieve more perceptual consistency in feature fusion. Experimental results on public IQA datasets show that the proposed RQFL-IQA is superior over existing methods.

Markov Blankets and Mirror Symmetries-Free Energy Minimization and Mesocortical Anatomy.

A theoretical account of development in mesocortical anatomy is derived from the free energy principle, operating in a neural field with both Hebbian and anti-Hebbian neural plasticity. An elementary structural unit is proposed, in which synaptic connections at mesoscale are arranged in paired patterns with mirror symmetry. Exchanges of synaptic flux in each pattern form coupled spatial eigenmodes, and the line of mirror reflection between the paired patterns operates as a Markov blanket, so that prediction errors in exchanges between the pairs are minimized. The theoretical analysis is then compared to the outcomes from a biological model of neocortical development, in which neuron precursors are selected by apoptosis for cell body and synaptic connections maximizing synchrony and also minimizing axonal length. It is shown that this model results in patterns of connection with the anticipated mirror symmetries, at micro-, meso- and inter-arial scales, among lateral connections, and in cortical depth. This explains the spatial organization and functional significance of neuron response preferences, and is compatible with the structural form of both columnar and noncolumnar cortex. Multi-way interactions of mirrored representations can provide a preliminary anatomically realistic model of cortical information processing.