ActInf GuestStream 076.2 ~ Anna Pereira: Active Inference Circle Book Project and Things Update
"Active Inference Circle Book Project and Things Update"
Anna Pereira
This talk looks to update the active inference for a fulfilling life book project as well foster collaboration and mutualism. The talk will provide a high level update of the book status, share some updates and experiences of the AII Research Fellows program, share strategy of how I'm using existing ecosystems outside AII to help foster additional growth, and include a conversation on satiated mutualism balanced with the necessity of realistic protective mechanisms. This talk will include a few slides, but be more conversational in format with a goal of creating transparency, inviting collaboration, and sharing out learnings that might help others in their journeys.
https://fellows.activeinference.institute/
Active Inference Institute information:
Website: https://activeinference.org/
Twitter: https://twitter.com/InferenceActive
Discord: https://discord.gg/8VNKNp4jtx
YouTube: https://www.youtube.com/c/ActiveInference/
Active Inference Livestreams: https://coda.io/@active-inference-institute/livestreams
CSID: f8284d7ad918e4ad
Content Managed by ContentSafe.co
27
views
1
comment
ActInf GuestStream 083.1 ~ "Embodied intelligence", Joshua Bongard
"Embodied intelligence"
Joshua Bongard
It is still laughably easy to foil AI with adversarial attacks. Why? Because such systems lack embodiment. But dropping deep learners into robots and calling the result Embodied AI misses the mark: Embodiment is about more than just having a body; it is about change. Consider how much the world changes from the perspective of a human as she grows from one cell into 10^13 of them. Grappling with such massive internal change makes grappling with external change, like learning to read or drive, easy by comparison. Thus, to realize safe AI, we must similarly create autonomous technologies whose internal physical changes pretrain them to handle external change, like new tasks or adversarial attacks. I’m excited to discuss a future filled with soft and biological robots that are capable of this “morphological pre-training”, and how it will make them useful, smart, and safe.
https://www.uvm.edu/cems/cs/profiles/josh_bongard
https://scholar.google.com/citations?user=Dj-kPasAAAAJ&hl=en
Xenobots, Biobots, and more!
Active Inference Institute information:
Website: https://activeinference.org/
Twitter: https://twitter.com/InferenceActive
Discord: https://discord.gg/8VNKNp4jtx
YouTube: https://www.youtube.com/c/ActiveInference/
Active Inference Livestreams: https://coda.io/@active-inference-institute/livestreams
CSID: cd30943d5f4794ce
Content Managed by ContentSafe.co
16
views
1
comment
MathArtStream 4 ~ Kirby Urner: “Dimension" in Synergetics
Kirby Urner: “Dimension" in Synergetics
https://www.grunch.net/4dsolutions/kirby.html
https://www.youtube.com/c/kirbyurner
https://github.com/4dsolutions/clarusway_data_analysis/blob/main/Kirby%20Notebooks/Intro_Bio.ipynb
https://coda.io/d/Math4Wisdom_d0SvdI3KSto/Synergetics-Videos_suOU9#_lu6H3
Active Inference Institute information:
Website: https://activeinference.org/
Twitter: https://twitter.com/InferenceActive
Discord: https://discord.gg/8VNKNp4jtx
YouTube: https://www.youtube.com/c/ActiveInference/
Active Inference Livestreams: https://coda.io/@active-inference-institute/livestreams
CSID: aa089476594d5d19
Content Managed by ContentSafe.co
16
views
Active Inference Insights 024 ~ Darius Parvizi-Wayne (Season Finale // Reflection)
In this lively and self-aware interview, Darius Parvizi-Wayne reflects on his podcast series "Active Inference Insights." We'll dive into the wonders of active inference, exploring how it helps us decode cognitive ecosystems and make sense of decision-making. Parvizi-Wayne shares the joys and challenges of turning dense theories into engaging discussions, revealing the podcast's success in making high-level concepts accessible. Expect some laughs as we unpack the series' impact on the research community, its knack for sparking interdisciplinary conversations, and its vision for the future. Join us for a fun yet insightful journey through the mind-expanding world of active inference.
Darius Parvizi-Wayne
https://scholar.google.com/citations?user=ahuMDH4AAAAJ&hl=en
https://x.com/dparviziwayne
Active Inference Institute information:
Website: https://activeinference.org/
Twitter: https://twitter.com/InferenceActive
Discord: https://discord.activeinference.institute/
YouTube: https://www.youtube.com/c/ActiveInference/
Active Inference Livestreams: https://video.activeinference.institute/
CSID: e91b8d2f09e5a2f1
Content Managed by ContentSafe.co
21
views
ActInf ArtStream 002.1 ~Aenne Brielmann: "Modelling individual aesthetic judgements over time"
"Modelling individual aesthetic judgements over time"
Aenne A. Brielmann, Max Berentelg and Peter Dayan
Published:18 December 2023 https://doi.org/10.1098/rstb.2022.0414
https://royalsocietypublishing.org/doi/full/10.1098/rstb.2022.0414
Abstract
Listening to music, watching a sunset—many sensory experiences are valuable to us, to a degree that differs significantly between individuals, and within an individual over time. We have theorized (Brielmann & Dayan 2022 Psychol. Rev. 129, 1319–1337 (doi:10.1037/rev0000337))) that these idiosyncratic values derive from the task of using experiences to tune the sensory-cognitive system to current and likely future input. We tested the theory using participants’ (n = 59) ratings of a set of dog images (n = 55) created using the NeuralCrossbreed morphing algorithm. A full realization of our model that uses feature representations extracted from image-recognizing deep neural nets (e.g. VGG-16) is able to capture liking judgements on a trial-by-trial basis (median r = 0.65), outperforming predictions based on population averages (median r = 0.01). Furthermore, the model’s learning component allows it to explain image sequence dependent rating changes, capturing on average 17% more variance in the ratings for the true trial order than for simulated random trial orders. This validation of our theory is the first step towards a comprehensive treatment of individual differences in evaluation.
This article is part of the theme issue ‘Art, aesthetics and predictive processing: theoretical and empirical perspectives’.
Active Inference Institute information:
Website: https://activeinference.org/
Twitter: https://twitter.com/InferenceActive
Discord: https://discord.gg/8VNKNp4jtx
YouTube: https://www.youtube.com/c/ActiveInference/
Active Inference Livestreams: https://coda.io/@active-inference-institute/livestreams
CSID: 108affd87abcd8ba
Content Managed by ContentSafe.co
24
views
Active Inference Institute ~ 2024 Quarterly Roundtable #2
Active Inference Institute information:
Website: https://activeinference.org/
Twitter: https://twitter.com/InferenceActive
Discord: https://discord.gg/8VNKNp4jtx
YouTube: https://www.youtube.com/c/ActiveInference/
Active Inference Livestreams: https://coda.io/@active-inference-institute/livestreams
CSID: bd553bfdc69a68ec
Content Managed by ContentSafe.co
10
views
ActInf ModelStream 012.1 ~ Fisher, Whyte, and Hohwy: An Active Inference Model of the Optimism Bias
"An Active Inference Model of the Optimism Bias"
Elizabeth L Fisher, Christopher J. Whyte, and Jakob Hohwy
https://osf.io/preprints/psyarxiv/xfb48
The optimism bias is a cognitive bias where individuals overestimate the likelihood of good outcomes and underestimate the likelihood of bad outcomes. Associated with improved quality of life, optimism bias is considered to be adaptive and is a promising avenue of research for mental health interventions in conditions where individuals lack optimism such as major depressive disorder. Here we lay the groundwork for future research on optimism as an intervention by introducing a domain general formal model of optimism bias, which can be applied in different task settings. Employing the active inference framework, we propose a model of the optimism bias as precision over a positive outcome. First, we simulate how optimism may be lost during development by exposure to negative events. We then ground our model in the empirical literature by showing how the developmentally acquired differences in optimism are expressed in a belief updating task typically used to assess optimism bias. Finally, we show how optimism affects action in a modified two-armed bandit task. Our model and the simulations it affords provide a computational basis for understanding how optimism bias may emerge, how it may be expressed in standard tasks used to assess optimism, and how it affects agents’ decision-making and actions; in combination, this provides a basis for future research on optimism as a mental health intervention.
https://doi.org/10.31234/osf.io/xfb48
Created: June 02, 2024
https://github.com/bethfisher-hub/optimism_simulations/
----
Active Inference Institute information:
Website: https://activeinference.org/
Twitter: https://twitter.com/InferenceActive
Discord: https://discord.gg/8VNKNp4jtx
YouTube: https://www.youtube.com/c/ActiveInference/
Active Inference Livestreams: https://coda.io/@active-inference-institute/livestreams
CSID: c411f7b08f93d3bf
Content Managed by ContentSafe.co
20
views
ActInf ModelStream 011.1 ~ Hadi Vafaii: "Poisson Variational Autoencoder"
"Poisson Variational Autoencoder"
Hadi Vafaii, Dekel Galor, Jacob L. Yates
https://arxiv.org/abs/2405.14473
[Submitted on 23 May 2024]
Variational autoencoders (VAE) employ Bayesian inference to interpret sensory inputs, mirroring processes that occur in primate vision across both ventral (Higgins et al., 2021) and dorsal (Vafaii et al., 2023) pathways. Despite their success, traditional VAEs rely on continuous latent variables, which deviates sharply from the discrete nature of biological neurons. Here, we developed the Poisson VAE (P-VAE), a novel architecture that combines principles of predictive coding with a VAE that encodes inputs into discrete spike counts. Combining Poisson-distributed latent variables with predictive coding introduces a metabolic cost term in the model loss function, suggesting a relationship with sparse coding which we verify empirically. Additionally, we analyze the geometry of learned representations, contrasting the P-VAE to alternative VAE models. We find that the P-VAEencodes its inputs in relatively higher dimensions, facilitating linear separability of categories in a downstream classification task with a much better (5x) sample efficiency. Our work provides an interpretable computational framework to study brain-like sensory processing and paves the way for a deeper understanding of perception as an inferential process.
----
Active Inference Institute information:
Website: https://activeinference.org/
Twitter: https://twitter.com/InferenceActive
Discord: https://discord.gg/8VNKNp4jtx
YouTube: https://www.youtube.com/c/ActiveInference/
Active Inference Livestreams: https://coda.io/@active-inference-institute/livestreams
CSID: e9aeb713e167ab71
Content Managed by ContentSafe.co
19
views
Active Inference LiveStream 057.2 ~ Active Data Selection and Information Seeking
Second participatory group discussion on the paper
Active Data Selection and Information Seeking
https://www.mdpi.com/1999-4893/17/3/118
Parr, Thomas, Karl Friston, and Peter Zeidman. 2024. "Active Data Selection and Information Seeking" Algorithms 17, no. 3: 118. https://doi.org/10.3390/a17030118
Abstract
Bayesian inference typically focuses upon two issues. The first is estimating the parameters of some model from data, and the second is quantifying the evidence for alternative hypotheses—formulated as alternative models. This paper focuses upon a third issue. Our interest is in the selection of data—either through sampling subsets of data from a large dataset or through optimising experimental design—based upon the models we have of how those data are generated. Optimising data-selection ensures we can achieve good inference with fewer data, saving on computational and experimental costs. This paper aims to unpack the principles of active sampling of data by drawing from neurobiological research on animal exploration and from the theory of optimal experimental design. We offer an overview of the salient points from these fields and illustrate their application in simple toy examples, ranging from function approximation with basis sets to inference about processes that evolve over time. Finally, we consider how this approach to data selection could be applied to the design of (Bayes-adaptive) clinical trials.
Keywords: experimental design; active sampling; information gain; Bayesian inference
Active Inference Institute information:
Website: https://activeinference.org/
Twitter: https://twitter.com/InferenceActive
Discord: https://discord.gg/8VNKNp4jtx
YouTube: https://www.youtube.com/c/ActiveInference/
Active Inference Livestreams: https://coda.io/@active-inference-institute/livestreams
CSID: 0b0cd1c41a875547
Content Managed by ContentSafe.co
11
views
ActInf GuestStream 084.1 ~ "The Nature of Habits and Agential Systems", Jesse G
The Nature of Habits and Agential Systems
Jesse G
Understanding persisting behavioral compositions, also known as habits and what they have to do with agent-based modeling of systems. Investigating agent-based modeling dynamics generally from the point of view of better understanding human behavior and provide insight on more optimal means to improve how we as humans manage our cognition and bodily functioning while facilitating sufficient symbiotic co-existence among each other as well as our environment and other agents as general intelligent multi-agent systems.
Active Inference Institute information:
Website: https://activeinference.org/
Twitter: https://twitter.com/InferenceActive
Discord: https://discord.gg/8VNKNp4jtx
YouTube: https://www.youtube.com/c/ActiveInference/
Active Inference Livestreams: https://coda.io/@active-inference-institute/livestreams
CSID: f3bab9b2be182084
Content Managed by ContentSafe.co
5
views
MathArtStream 3 ~ "Categorical Mental Imagery: Visualizing the 4th Spatial Dimension" with Jim Zhon
"Categorical Mental Imagery: Visualizing the 4th Spatial Dimension"
Jim Zhong, Shanna Dobson
Active Inference Institute information:
Website: https://activeinference.org/
Twitter: https://twitter.com/InferenceActive
Discord: https://discord.gg/8VNKNp4jtx
YouTube: https://www.youtube.com/c/ActiveInference/
Active Inference Livestreams: https://coda.io/@active-inference-institute/livestreams
CSID: c04b59d251a5c7f4
Content Managed by ContentSafe.co
2
views
Active Inference LiveStream 057.1 ~ Active Data Selection and Information Seeking
First participatory group discussion on the paper.
(See #057.0 background & context video https://www.youtube.com/live/GoXtkxLGkjQ )
Active Data Selection and Information Seeking
https://www.mdpi.com/1999-4893/17/3/118
Parr, Thomas, Karl Friston, and Peter Zeidman. 2024. "Active Data Selection and Information Seeking" Algorithms 17, no. 3: 118. https://doi.org/10.3390/a17030118
Abstract
Bayesian inference typically focuses upon two issues. The first is estimating the parameters of some model from data, and the second is quantifying the evidence for alternative hypotheses—formulated as alternative models. This paper focuses upon a third issue. Our interest is in the selection of data—either through sampling subsets of data from a large dataset or through optimising experimental design—based upon the models we have of how those data are generated. Optimising data-selection ensures we can achieve good inference with fewer data, saving on computational and experimental costs. This paper aims to unpack the principles of active sampling of data by drawing from neurobiological research on animal exploration and from the theory of optimal experimental design. We offer an overview of the salient points from these fields and illustrate their application in simple toy examples, ranging from function approximation with basis sets to inference about processes that evolve over time. Finally, we consider how this approach to data selection could be applied to the design of (Bayes-adaptive) clinical trials.
Keywords: experimental design; active sampling; information gain; Bayesian inference
Active Inference Institute information:
Website: https://activeinference.org/
Twitter: https://twitter.com/InferenceActive
Discord: https://discord.gg/8VNKNp4jtx
YouTube: https://www.youtube.com/c/ActiveInference/
Active Inference Livestreams: https://coda.io/@active-inference-institute/livestreams
CSID: 996dfcf8415c8a8b
Content Managed by ContentSafe.co
6
views
Active Inference LiveStream 057.0 ~ Active Data Selection and Information Seeking
Background and Context video for:
Active Data Selection and Information Seeking
https://www.mdpi.com/1999-4893/17/3/118
Parr, Thomas, Karl Friston, and Peter Zeidman. 2024. "Active Data Selection and Information Seeking" Algorithms 17, no. 3: 118. https://doi.org/10.3390/a17030118
Abstract
Bayesian inference typically focuses upon two issues. The first is estimating the parameters of some model from data, and the second is quantifying the evidence for alternative hypotheses—formulated as alternative models. This paper focuses upon a third issue. Our interest is in the selection of data—either through sampling subsets of data from a large dataset or through optimising experimental design—based upon the models we have of how those data are generated. Optimising data-selection ensures we can achieve good inference with fewer data, saving on computational and experimental costs. This paper aims to unpack the principles of active sampling of data by drawing from neurobiological research on animal exploration and from the theory of optimal experimental design. We offer an overview of the salient points from these fields and illustrate their application in simple toy examples, ranging from function approximation with basis sets to inference about processes that evolve over time. Finally, we consider how this approach to data selection could be applied to the design of (Bayes-adaptive) clinical trials.
Keywords: experimental design; active sampling; information gain; Bayesian inference
Active Inference Institute information:
Website: https://activeinference.org/
Twitter: https://twitter.com/InferenceActive
Discord: https://discord.gg/8VNKNp4jtx
YouTube: https://www.youtube.com/c/ActiveInference/
Active Inference Livestreams: https://coda.io/@active-inference-institute/livestreams
CSID: cb14998e47baaa9e
Content Managed by ContentSafe.co
6
views
MathArtStream 2 ~ "Towards Tractable Insights: Dynamics and Immersion in, out and between MathArt"
"Towards Tractable Insights: Dynamics and Immersion in, out and between MathArt"
Shanna Dobson, Dean Tickles, Stephen Sillett
Active Inference Institute information:
Website: https://activeinference.org/
Twitter: https://twitter.com/InferenceActive
Discord: https://discord.gg/8VNKNp4jtx
YouTube: https://www.youtube.com/c/ActiveInference/
Active Inference Livestreams: https://coda.io/@active-inference-institute/livestreams
CSID: 442394636fc59da9
Content Managed by ContentSafe.co
4
views
MathArtStream 1 ~ "Introducing MathArt Conversations"
"Introducing MathArt Conversations: Scaffolding Apostles of New Thinking"
Shanna Dobson, Héctor Manrique, Daniel Friedman
Active Inference Institute information:
Website: https://activeinference.org/
Twitter: https://twitter.com/InferenceActive
Discord: https://discord.gg/8VNKNp4jtx
YouTube: https://www.youtube.com/c/ActiveInference/
Active Inference Livestreams: https://coda.io/@active-inference-institute/livestreams
CSID: 0b7ffe421c37092e
Content Managed by ContentSafe.co
4
views
ActInf GuestStream 083.1 ~ Simon Wardley
Simon Wardley is the creator of Wardley Mapping, an open-source collaborative strategic mapping technique that has gained viral adoption in the business and technology communities. At its core, it creates space to eloquently capture collective intelligence.
Wardley Mapping helps groups and organizations develop situational awareness by visually representing patterns and trends in the evolution of technologies, capabilities, and user needs within a competitive landscape. His pioneering work on Wardley Mapping has been influential in understanding complex, evolving systems by applying core principles of pattern recognition and strategic adaptation. It offers a practice of exploring collective intelligence, and how landscapes are shaped beyond theory and speculation.
We’ll explore his experiences, insights and surprises as a global trendsetter facilitating hundreds of diverse groups working on a wide range of topics, including the future of education, work, and technology.
Simon's work captures mega-trends that reveal how market forces shape divergence and consensus that manifest in powerful, unusual and fascinating ways.
Learn more..
https://medium.com/wardleymaps/on-being-lost-2ef5f05eb1ec
https://learnwardleymapping.com/book/
https://www.linkedin.com/in/simonwardley/
https://github.com/wardley-maps-community/awesome-wardley-maps
Active Inference Institute information:
Website: https://activeinference.org/
Twitter: https://twitter.com/InferenceActive
Discord: https://discord.gg/8VNKNp4jtx
YouTube: https://www.youtube.com/c/ActiveInference/
Active Inference Livestreams: https://coda.io/@active-inference-institute/livestreams
CSID: 36e562b62b920481
Content Managed by ContentSafe.co
2
views
ActInf GuestStream 080.1 ~ Laura Desirée Di Paolo: "Active Inference Goes to School"
"Active Inference Goes to School. The Importance of Active Learning in the Age of Large Language Models"
Laura Desirée Di Paolo, Ben White, Avel GUÉNIN--CARLUT, Axel Constant, and Andy Clark
https://osf.io/preprints/osf/zwa83
Human learning essentially involves embodied interactions with the material world. But our worlds now include increasing numbers of powerful and (apparently) disembodied generative AIs. In what follows we ask how best to understand these new (somewhat “alien”, because of their disembodied nature) resources and how to incorporate them in our educational practices. We focus on methodologies that encourage exploration and embodied interactions with ‘prepared’ material environments, such as the carefully organised settings of Montessori education. Using the Active Inference Framework, we approach our questions by thinking about human learning as epistemic foraging and prediction error minimization. We end by arguing that generative AIs should figure naturally as new elements in prepared learning environments by facilitating sequences of precise prediction error enabling trajectories of self-correction. In these ways we anticipate new synergies between (apparently) disembodied and (essentially) embodied forms of intelligence.
Active Inference Institute information:
Website: https://activeinference.org/
Twitter: https://twitter.com/InferenceActive
Discord: https://discord.gg/8VNKNp4jtx
YouTube: https://www.youtube.com/c/ActiveInference/
Active Inference Livestreams: https://coda.io/@active-inference-institute/livestreams
CSID: 31daffdf373d2e61
Content Managed by ContentSafe.co
4
views
ActInf GuestStream 082.1 ~ Robert Worden "Bayesian Model-Based Cognition: The Requirement Equation"
"Bayesian Model-Based Cognition: The Requirement Equation"
Robert Worden
Active Inference Institute information:
Website: https://activeinference.org/
Twitter: https://twitter.com/InferenceActive
Discord: https://discord.gg/8VNKNp4jtx
YouTube: https://www.youtube.com/c/ActiveInference/
Active Inference Livestreams: https://coda.io/@active-inference-institute/livestreams
CSID: 2e9de98c977e4d4e
Content Managed by ContentSafe.co
1
view
ActInf ArtStream 001.1 ~ "Order and change in art", Sander Van de Cruys, Jacopo Frascaroli
"Order and change in art: towards an active inference account of aesthetic experience"
Sander Van de Cruys, Jacopo Frascaroli and Karl Friston
Published:18 December 2023 https://doi.org/10.1098/rstb.2022.0411
https://royalsocietypublishing.org/doi/10.1098/rstb.2022.0411
How to account for the power that art holds over us? Why do artworks touch us deeply, consoling, transforming or invigorating us in the process? In this paper, we argue that an answer to this question might emerge from a fecund framework in cognitive science known as predictive processing (a.k.a. active inference). We unpack how this approach connects sense-making and aesthetic experiences through the idea of an ‘epistemic arc’, consisting of three parts (curiosity, epistemic action and aha experiences), which we cast as aspects of active inference. We then show how epistemic arcs are built and sustained by artworks to provide us with those satisfying experiences that we tend to call ‘aesthetic’. Next, we defuse two key objections to this approach; namely, that it places undue emphasis on the cognitive component of our aesthetic encounters—at the expense of affective aspects—and on closure and uncertainty minimization (order)—at the expense of openness and lingering uncertainty (change). We show that the approach offers crucial resources to account for the open-ended, free and playful behaviour inherent in aesthetic experiences. The upshot is a promising but deflationary approach, both philosophically informed and psychologically sound, that opens new empirical avenues for understanding our aesthetic encounters.
Active Inference Institute information:
Website: https://activeinference.org/
Twitter: https://twitter.com/InferenceActive
Discord: https://discord.gg/8VNKNp4jtx
YouTube: https://www.youtube.com/c/ActiveInference/
Active Inference Livestreams: https://coda.io/@active-inference-institute/livestreams
CSID: b87e851199bc9a7d
Content Managed by ContentSafe.co
2
views
ActInf OrgStream 008.1 ~ Nathan Schneider: "Governable Spaces: Democratic Design for Online Life"
Governable Spaces
Democratic Design for Online Life
by Nathan Schneider (Author), Darija Medic (Illustrator)
https://www.ucpress.edu/book/9780520393943/governable-spaces
When was the last time you participated in an election for an online group chat or sat on a jury for a dispute about a controversial post? Platforms nudge users to tolerate nearly all-powerful admins, moderators, and "benevolent dictators for life." In Governable Spaces, Nathan Schneider argues that the internet has been plagued by a phenomenon he calls "implicit feudalism": a bias, both cultural and technical, for building communities as fiefdoms. The consequences of this arrangement matter far beyond online spaces themselves, as feudal defaults train us to give up on our communities' democratic potential, inclining us to be more tolerant of autocratic tech CEOs and authoritarian tendencies among politicians. But online spaces could be sites of a creative, radical, and democratic renaissance. Using media archaeology, political theory, and participant observation, Schneider shows how the internet can learn from governance legacies of the past to become a more democratic medium, responsive and inventive unlike anything that has come before.
Active Inference Institute information:
Website: https://activeinference.org/
Twitter: https://twitter.com/InferenceActive
Discord: https://discord.gg/8VNKNp4jtx
YouTube: https://www.youtube.com/c/ActiveInference/
Active Inference Livestreams: https://coda.io/@active-inference-institute/livestreams
CSID: 3d0d669e7f7388f6
Content Managed by ContentSafe.co
3
views
ActInf ModelStream 010.1 ~ Building a Drone with RxInfer.jl ~ Bart van Erp, Albert Podusenko
Witness the power of message passing as we guide you through the process of building a drone in real-time using our toolbox RxInfer.jl. Whether you're a novice intrigued by neuroscience or a seasoned researcher, this immersive demonstration promises to unveil the practical applications of active inference, revolutionizing our approach to autonomous systems.
https://github.com/ReactiveBayes/RxInfer.jl
https://rxinfer.ml/
https://lazydynamics.com/
----
Active Inference Institute information:
Website: https://activeinference.org/
Twitter: https://twitter.com/InferenceActive
Discord: https://discord.gg/8VNKNp4jtx
YouTube: https://www.youtube.com/c/ActiveInference/
Active Inference Livestreams: https://coda.io/@active-inference-institute/livestreams
CSID: 9c601b76db4f88f8
Content Managed by ContentSafe.co
2
views
ActInf GuestStream 079.1 ~ Ryota Kanai: "Meta-Representations as Representations of Processes"
"Meta-Representations as Representations of Processes"
Ryota Kanai, Ryota Takatsuki, and Ippei Fujisawa
https://osf.io/preprints/psyarxiv/zg27u
In this study, we explore how the notion of meta-representations in Higher-Order Theories (HOT) of consciousness can be implemented in computational models. HOT suggests that consciousness emerges from meta-representations, which are representations of first-order sensory representations. However, translating this abstract concept into a concrete computational model, such as those used in artificial intelligence, presents a theoretical challenge. For example, a simplistic interpretation of meta-representation as a representation of representation makes the notion rather trivial and ubiquitous. Here, we propose a refined interpretation of meta-representations. Contrary to the simplistic view of meta-representations as mere transformations of the first-order representational states or confidence estimates, we argue that meta-representations are representations of the processes that generate first-order representations. This presents a process-oriented view whereby meta-representations capture the qualitative aspect of how sensory information is transformed into first-order representations. To concretely illustrate and operationalize thus formulated notion of meta-representation, we constructed "meta-networks" designed to explicitly model meta-representations within deep learning architectures. Specifically, we constructed meta-networks by implementing autoencoders of first-order neural networks. In this architecture, the latent spaces embedding those first-order networks correspond to the meta-representations of first-order networks. By applying meta-networks to embed neural networks trained to encode visual and auditory datasets, we show that the meta-representations of first-order networks successfully capture the qualitative aspects of those networks by separating the visual and auditory networks in the meta-representation space. We argue that such meta-representations would be useful for quantitatively compare and contrast the qualitative differences of computational processes. While whether such meta-representational systems exist in the human brain remains an open question, this formulation of meta-representation offers a new empirically testable hypothesis that there are brain regions that represent the processes of transforming a representation in one brain region to a representation in another brain region. Furthermore, this form of meta-representations might underlie our ability to describe the qualitative aspect of sensory experience or qualia.
Active Inference Institute information:
Website: https://activeinference.org/
Twitter: https://twitter.com/InferenceActive
Discord: https://discord.gg/8VNKNp4jtx
YouTube: https://www.youtube.com/c/ActiveInference/
Active Inference Livestreams: https://coda.io/@active-inference-institute/livestreams
CSID: ebb7a10035fed94d
Content Managed by ContentSafe.co
6
views
ActInf GuestStream 078.1 ~ Juan Diego Bogotá: "What could come before time?"
Juan Diego Bogotá
"What could come before time? Intertwining affectivity and temporality at the basis of intentionality"
https://link.springer.com/article/10.1007/s11097-024-09973-y
The enactive approach to cognition and the phenomenological tradition have in common a wide conception of ‘intentionality’. Within these frameworks, intentionality is understood as a general openness to the world. For classical phenomenologists, the most basic subjective structure that allows for such openness is time-consciousness. Some enactivists, while inspired by the phenomenological tradition, have nevertheless argued that affectivity is more basic, being that which gives rise to the temporal flow of consciousness. In this paper, I assess the relationship between temporality and affectivity from both a phenomenological and an enactive perspective. I argue that, as opposed to the classical phenomenological view (which favours temporality), and to the enactive view (which favours affectivity), we must take affectivity and temporality as co-emergent. Jointly, affectivity and temporality constitute the basic structures of intentionality. Additionally, using examples from phenomenological psychopathology, I conclude that all intentionality is defined by an anticipatory and affective structure that gives rise to general feelings related to our bodily possibilities in the world.
Active Inference Institute information:
Website: https://activeinference.org/
Twitter: https://twitter.com/InferenceActive
Discord: https://discord.gg/8VNKNp4jtx
YouTube: https://www.youtube.com/c/ActiveInference/
Active Inference Livestreams: https://coda.io/@active-inference-institute/livestreams
CSID: 7377a165134b4ae7
Content Managed by ContentSafe.co
4
views
ActInf MathStream 010.1 ~ Thomas Varley "Generalized decomposition of multivariate information"
Generalized decomposition of multivariate information
Thomas Varley
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10843128/
Since its introduction, the partial information decomposition (PID) has emerged as a powerful, information-theoretic technique useful for studying the structure of (potentially higher-order) interactions in complex systems. Despite its utility, the applicability of the PID is restricted by the need to assign elements as either “sources” or “targets”, as well as the specific structure of the mutual information itself. Here, I introduce a generalized information decomposition that relaxes the source/target distinction while still satisfying the basic intuitions about information. This approach is based on the decomposition of the Kullback-Leibler divergence, and consequently allows for the analysis of any information gained when updating from an arbitrary prior to an arbitrary posterior. As a result, any information-theoretic measure that can be written as a linear combination of Kullback-Leibler divergences admits a decomposition in the style of Williams and Beer, including the total correlation, the negentropy, and the mutual information as special cases. This paper explores how the generalized information decomposition can reveal novel insights into existing measures, as well as the nature of higher-order synergies. We show that synergistic information is intimately related to the well-known Tononi-Sporns-Edelman (TSE) complexity, and that synergistic information requires a similar integration/segregation balance as a high TSE complexity. Finally, I end with a discussion of how this approach fits into other attempts to generalize the PID and the possibilities for empirical applications.
Active Inference Institute information:
Website: https://activeinference.org/
Twitter: https://twitter.com/InferenceActive
Discord: https://discord.gg/8VNKNp4jtx
YouTube: https://www.youtube.com/c/ActiveInference/
Active Inference Livestreams: https://coda.io/@active-inference-institute/livestreams
CSID: ccd5b36fe12011c2
Content Managed by ContentSafe.co
8
views
ActInf GuestStream 075.1 ~ Matsumura et al.: "Active Inference With Empathy Mechanism""
Active Inference With Empathy Mechanism for Socially Behaved Artificial Agents in Diverse Situations
Tadayuki Matsumura, Hiroyuki Mizuno, Kanako Esaki
松村忠幸, 水野弘之, 江﨑佳奈子
https://direct.mit.edu/artl/article/d...
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.
Active Inference Institute information:
Website: https://activeinference.org/
Twitter: https://twitter.com/InferenceActive
Discord: https://discord.gg/8VNKNp4jtx
YouTube: https://www.youtube.com/c/ActiveInfer...
Active Inference Livestreams: https://coda.io/@active-inference-ins...
2
views