The Fully Bayesian Significance Test (FBST) is a coherent Bayesian significance test for sharp hypotheses. This paper proposes the FBST as a model selection tool for general mixture models, and compares its performance with Mclust, a model-based clustering software. The FBST robust performance strongly encourages further developments and investigations.

Gene clustering is a useful exploratory technique to group together genes with similar expression levels under distinct cell cycle phases or distinct conditions. It helps the biologist to identify potentially meaningful relationships between genes. In this study, we propose a clustering method based on multivariate normal mixture models, where the number of clusters is predicted via sequential hypothesis tests: at each step, the method considers a mixture model of m components (m = 2 in the first step) (...) and tests if in fact it should be m - 1. If the hypothesis is rejected, m is increased and a new test is carried out. The method continues (increasing m) until the hypothesis is accepted. The theoretical core of the method is the full Bayesian significance test, an intuitive Bayesian approach, which needs no model complexity penalization nor positive probabilities for sharp hypotheses. Numerical experiments were based on a cDNA microarray dataset consisting of expression levels of 205 genes belonging to four functional categories, for 10 distinct strains of Saccharomyces cerevisiae. To analyze the method’s sensitivity to data dimension, we performed principal components analysis on the original dataset and predicted the number of classes using 2 to 10 principal components. Compared to Mclust (model-based clustering), our method shows more consistent results. (shrink)

This article describes the Full Bayesian Significance Test for the problem of separate hypotheses. Numerical experiments are performed for the Gompertz vs. Weibull life span test.

For Anaxagoras, both before the beginning of the world and in the present, “all is together” and “everything is in everything.” Various modern interpretations abound regarding the identity of this “mixture.” It has been explained as an aggregation of particles or as a continuous “fusion” of different sorts of ingredients. However—even though they are not usually recognized as a distinct group—there are a number of other scholars who, without seemingly knowing each other, have offered a different interpreta- tion: Anaxagoras’ (...) mixture as an “interpenetration” of different ingredients, which are as far-extended as the whole mixture is. As a result, there are different entities occupying the same place at the same time. This explanation assigns to Anaxagoras the same model of mixture which was later used by the Stoics. A new book by Marmodoro helps us to clarify this position. (shrink)

The data analyzed in this paper are part of the results described in Bueno et al. (2000). Three cytogenetics endpoints were analyzed in three populations of a species of wild rodent – Akodon montensis – living in an industrial, an agricultural, and a preservation area at the Itajaí Valley, State of Santa Catarina, Brazil. The polychromatic/normochromatic ratio, the mitotic index, and the frequency of micronucleated polychromatic erythrocites were used in an attempt to establish a genotoxic profile of each area. It (...) was assumed that the three populations were in the same conditions with respect to the influence of confounding factors such as animal age, health, nutrition status, presence of pathogens, and intra- and inter-populational genetic variability. Therefore, any differences found in the endpoints analyzed could be attributed to the external agents present in each area. The statistical models used in this paper are mixtures of negative-binomials and Poisson variables. The Poisson variables are used as approximations of binomials for rare events. The mixing distributions are beta densities. The statistical analyzes are under the bayesian perspective, as opposed to the frequentist ones often considered in the literature, as for instance in Bueno et al. (2000). (shrink)

The Fully Bayesian Significance Test (FBST) is a coherent Bayesian significance test for sharp hypotheses. This paper explores the FBST as a model selection tool for general mixture models, and gives some computational experiments for Multinomial-Dirichlet-Normal-Wishart models.

We review some results in the theory of non-relativistic quantum unstable systems. We account for the most important definitions of quantum resonances that we identify with unstable quantum systems. Then, we recall the properties and construction of Gamow states as vectors in some extensions of Hilbert spaces, called Rigged Hilbert Spaces. Gamow states account for the purely exponential decaying part of a resonance; the experimental exponential decay for long periods of time physically characterizes a resonance. We briefly discuss one of (...) the most usual models for resonances: the Friedrichs model. Using an algebraic formalism for states and observables, we show that Gamow states cannot be pure states or mixtures from a standard view point. We discuss some additional properties of Gamow states, such as the possibility of obtaining mean values of certain observables on Gamow states. A modification of the time evolution law for the linear space spanned by Gamow shows that some non-commuting observables on this space become commuting for large values of time. We apply Gamow states for a possible explanation of the Loschmidt echo. (shrink)

An important problem with machine learning is that when label number n>2, it is very difficult to construct and optimize a group of learning functions, and we wish that optimized learning functions are still useful when prior distribution P(x) (where x is an instance) is changed. To resolve this problem, the semantic information G theory, Logical Bayesian Inference (LBI), and a group of Channel Matching (CM) algorithms together form a systematic solution. MultilabelMultilabel A semantic channel in the G theory consists (...) of a group of truth functions or membership functions. In comparison with likelihood functions, Bayesian posteriors, and Logistic functions used by popular methods, membership functions can be more conveniently used as learning functions without the above problem. In Logical Bayesian Inference (LBI), every label’s learning is independent. For Multilabel learning, we can directly obtain a group of optimized membership functions from a big enough sample with labels, without preparing different samples for different labels. A group of Channel Matching (CM) algorithms are developed for machine learning. For the Maximum Mutual Information (MMI) classification of three classes with Gaussian distributions on a two-dimensional feature space, 2-3 iterations can make mutual information between three classes and three labels surpass 99% of the MMI for most initial partitions. For mixture models, the Expectation-Maxmization (EM) algorithm is improved and becomes the CM-EM algorithm, which can outperform the EM algorithm when mixture ratios are imbalanced, or local convergence exists. The CM iteration algorithm needs to combine neural networks for MMI classifications on high-dimensional feature spaces. LBI needs further studies for the unification of statistics and logic. (shrink)

It is proposed that our memory is made up of individual, unconnected Records, to each of which is attached a Heading. Retrieval of a Record can only be accomplished by addressing the attached Heading, the contents of which cannot itself be retrieved. Each Heading is made up of a mixture of content in more or less literal form and context, the latter including specification of environment and of internal states (e.g. drug states and mood). This view of memory allows (...) an easy account of a number of natural memory phenomena as well as a variety of laboratory findings such as the differences between recall and recognition. The theory further proposes that Headed Records can neither be deleted nor modified. Data apparently against such a hypothesis can be accounted for in terms of the retrieval process. (shrink)

This chapter from Michel Serres’s comprehensive study on Leibniz—"The System of Leibniz and its Mathematical Models [Le système de Leibniz et ses modèles mathématiques]"—examines Leibniz’s epistemological framework. This framework, which Leibniz developed for a large part in his “Meditations on Knowledge, Truth, and Ideas [Meditationes de cognitione, veritate et ideis],” is pitched against Descartes’s "Meditations on First Philosophy [Meditationes de Prima Philosophia]" and the method of systematic doubt developed therein. While Descartes rejects any knowledge with the slightest possibility of (...) falsehood, Leibniz accepts knowledge with even a minimal degree of truth. Leibniz’s approach involves a progressive genesis of truth through a series of filters, each further refining knowledge. This process is intrinsically linked to infinitism and combinatorics, allowing for a gradual differentiation of the distinct from the indistinct. The filters are applied sequentially, with each filter proceeding through its own oppositions, resulting in a spectrum of knowledge ranging from obscure to clear, confused to distinct, inadequate to adequate, and symbolic to intuitive. This framework facilitates pluralism within rationalism that proceeds through evolving or regional truths. The filters overlap, and only the final stage of knowledge is devoid of mixture. Leibniz’s epistemology is non-Cartesian, as it relativizes truth and falsehood at each stage of the filtering process. It also accounts for the limitations of sensory perception, attributing them to the imperfection of the human mind in its current state. (Translated by Martijn Boven). (shrink)

This article presents important properties of standard discrete distributions and its conjugate densities. The Bernoulli and Poisson processes are described as generators of such discrete models. A characterization of distributions by mixtures is also introduced. This article adopts a novel singular notation and representation. Singular representations are unusual in statistical texts. Nevertheless, the singular notation makes it simpler to extend and generalize theoretical results and greatly facilitates numerical and computational implementation.

In this paper, a novel experimental task is developed for testing the highly influential, but experimentally underexplored, possible worlds account of conditionals (Stalnaker, 1968; Lewis, 1973). In Experiment 1, this new task is used to test both indicative and subjunctive conditionals. For indicative conditionals, five competing truth tables are compared, including the previously untested, multi-dimensional possible worlds semantics of Bradley (2012). In Experiment 2, these results are replicated and it is shown that they cannot be accounted for by an alternative (...) hypothesis proposed by our reviewers. In Experiment 3, individual variation in truth assignments of indicative conditionals is investigated via Bayesian mixture models that classify participants as following one of several competing models. As a novelty of this study, it is found that a possible worlds semantics of Lewis and Stalnaker is capable of accounting for participants’ aggregate truth value assignments in this task. Applied to indicative conditionals, we show across three experiments, that the theory both captures participants’ truth values at the aggregate level (Experiments 1 and 2) and that it makes up the largest subgroup in the analysis of individual variation in our experimental paradigm (Experiment 3). (shrink)

In this manuscript we study individual variation in the interpretation of conditionals by establishing individual profiles of the participants based on their behavioral responses and reflective attitudes. To investigate the participants’ reflective attitudes we introduce a new experimental paradigm called the Scorekeeping Task, and a Bayesian mixture model tailored to analyze the data. The goal is thereby to identify the participants who follow the Suppositional Theory of conditionals and Inferentialism and to investigate their performance on the uncertain and-to-if inference (...) task. (shrink)

This study analyzed the behavioral characteristics of ICT users among postgraduate students leveraging the Latent Class Analysis (LCA). The study, anchored on the Planned Behavior Theory, followed the exploratory research design. It adopted the cluster random sampling technique in selecting 1,023 respondents from a population of 2,923 postgraduate students in four federal universities in South-South Nigeria. “Behavioural Characteristics and Job Creation Questionnaire (BCJCQ),” developed by the researchers, was used for data collection. Upon data collection and LCA analysis, the five-class solution (...) was accepted as the best-fitting model, based on statistical fit indicators (such as AIC, BIC, entropy, Gsq, and Chsq) and theoretical grounds. Consequently, five classes of behavioral ICT users were identified and named based on their item–response probability, conditional on class. The five classes were named Trendy, Outmoded, Pragmatic, Disciplined, and Social users of ICT, with their unique characteristics discussed. The study tested for job creation differences among the classes using a one-way ANOVA and found a significant difference. On average, pragmatic users of ICT created more jobs than social, disciplined, and outmoded users. Trendy users were, on average, the minor job-creating class of ICT users. The study compared the bivariate differences in job creation among the classes using the Tukey HSD test of multiple pairwise comparisons. Based on the results obtained, discussions were made with implications for further research in the evolving area of LCA. (shrink)

A new trend in deep learning, represented by Mutual Information Neural Estimation (MINE) and Information Noise Contrast Estimation (InfoNCE), is emerging. In this trend, similarity functions and Estimated Mutual Information (EMI) are used as learning and objective functions. Coincidentally, EMI is essentially the same as Semantic Mutual Information (SeMI) proposed by the author 30 years ago. This paper first reviews the evolutionary histories of semantic information measures and learning functions. Then, it briefly introduces the author’s semantic information G theory with (...) the rate-fidelity function R(G) (G denotes SeMI, and R(G) extends R(D)) and its applications to multi-label learning, the maximum Mutual Information (MI) classification, and mixture models. Then it discusses how we should understand the relationship between SeMI and Shannon’s MI, two generalized entropies (fuzzy entropy and coverage entropy), Autoencoders, Gibbs distributions, and partition functions from the perspective of the R(G) function or the G theory. An important conclusion is that mixture models and Restricted Boltzmann Machines converge because SeMI is maximized, and Shannon’s MI is minimized, making information efficiency G/R close to 1. A potential opportunity is to simplify deep learning by using Gaussian channel mixture models for pre-training deep neural networks’ latent layers without considering gradients. It also discusses how the SeMI measure is used as the reward function (reflecting purposiveness) for reinforcement learning. The G theory helps interpret deep learning but is far from enough. Combining semantic information theory and deep learning will accelerate their development. (shrink)

This paper is a tentative step towards a historical cognitive science, in the domain of memory and personal identity. I treat theoretical models of memory in history as specimens of the way cultural norms and artifacts can permeate ('proto')scientific views of inner processes. I apply this analysis to the topic of psychological control over one's own body, brain, and mind. Some metaphors and models for memory and mental representation signal the projection inside of external aids. Overtly at least, (...) medieval and Renaissance theorists agreed that such models had to allow for, or even guarantee, some conception of cognitive order and discipline. Individual memory traces had to be independent, not mixed up or interfering with others. The long tradition of improving or bypassing 'natural memory' by deliberately internalizing artefactual models was part of an arduous process of self-fashioning. Moral panic about confusion and mixture features centrally in the imposition of cognitive discipline in local memory traditions. (shrink)

Contemporary biology has inherited two key assumptions from the Modern Synthesis about the nature of population lineages: sexual reproduction is the exemplar for how individuals in population lineages inherit traits from their parents, and random mating is the exemplar for reproductive interaction. While these assumptions have been extremely fruitful for a number of fields, such as population genetics and phylogenetics, they are increasingly unviable for studying the full diversity and evolution of life. I introduce the “mixture” account of population (...) lineages that escapes these assumptions by dissolving the Modern Synthesis’s sharp line separating reproduction and development and characterizing reproductive integration in population lineages by the ephemerality of isolated subgroups rather than random mating. The mixture account provides a single criterion for reproductive integration that accommodates both sexual and asexual reproduction, unifying their treatment under Kevin de Queiroz’s generalized lineage concept of species. The account also provides a new basis for empirically assessing the effect of random mating as an idealization on the empirical adequacy of population genetic models. (shrink)

I defend a one category ontology: an ontology that denies that we need more than one fundamental category to support the ontological structure of the world. Categorical fundamentality is understood in terms of the metaphysically prior, as that in which everything else in the world consists. One category ontologies are deeply appealing, because their ontological simplicity gives them an unmatched elegance and spareness. I’m a fan of a one category ontology that collapses the distinction between particular and property, replacing it (...) with a single fundamental category of intrinsic characters or qualities. We may describe the qualities as qualitative charactersor as modes, perhaps on the model of Aristotelian qualitative (nonsubstantial) kinds, and I will use the term “properties” interchangeably with “qualities”. The qualities are repeatable and reasonably sparse, although, as I discuss in section 2.6, there are empirical reasons that may suggest, depending on one’s preferred fundamental physical theory, that they include irreducibly intensive qualities. There are no uninstantiated qualities. I also assume that the fundamental qualitative natures are intrinsic, although physics may ultimately suggest that some of them are extrinsic. On my view, matter, concrete objects, abstract objects, and perhaps even spacetime are constructed from mereological fusions of qualities, so the world is simply a vast mixture of qualities, including polyadic properties (i.e., relations). This means that everything there is, including concrete objects like persons or stars, is a quality, a qualitative fusion, or a portion of the extended qualitative fusion that is the worldwhole. I call my view mereological bundle theory. (shrink)

The aim of this text, derived from an investigation, is to propose a reflection on the future of the university, from the problematization of the neoliberal rationality impact on the forms of knowledge production and organization of academic life. To develop this analysis, the main transformations of the modern university since its advent with the founding of the Berlin University by Wilhelm von Humboldt, in 1810, who created the model of the research university, until the emergence of the neoliberal university (...) model. The purpose of the article is to highlight the rationalities that govern the transformation of the university over time. This historical analysis attempts to demonstrate that no model can be considered natural, absolute, inevitable, non-negotiable, and that and that the university as an institution is inhabited by a mixture of contradictory rationalities. To conclude, the Derrida's essay (2001), is used to show that the uncontested adoption of the neoliberal model can destroy the autonomy of research and teaching that until today characterized the modern university, and compromise its mission in the future. (shrink)

There is a commonly made distinction between two types of scientists: risk-taking, trailblazing mavericks and detail-oriented followers. A number of recent papers have discussed the question what a desirable mixture of mavericks and followers looks like. Answering this question is most useful if a scientific community can be steered toward such a desirable mixture. One attractive route is through credit incentives: manipulating rewards so that reward-seeking scientists are likely to form the desired mixture of their own accord. (...) Here I argue that this idea is less straightforward than it may seem. Interpreting mavericks as scientists who prioritize rewards over speed and risk, I show in a deliberatively simple model that there is a fixed mixture which is not particularly likely to be desirable and which credit incentives cannot alter. I consider a way around this result, but this has some major drawbacks. I conclude that credit incentives are not as promising a way to create a desirable mixture of mavericks and followers as one might have thought. (shrink)

In this paper I offer a new interpretation of Stoic ontology. I aim to explain the nature of, and relations between, (i) the fundamental items of their physics, bodies; (ii) the incorporeal items about which they theorized no less; and (iii) universals, towards which the Stoic attitude seems to be a bizarre mixture of realism and anti-realism. In the first half of the paper I provide a new model to explain the relationship between those items in (i) and (ii). (...) This model clears up several problems in Stoicism and gives a precise answer to the question of how bodies and incorporeals differ in their mode of being. The second half of the paper considers the vexed issue of the status of those items in (iii), Stoic universals. (shrink)

Abstract: Heart diseases are increasing daily at a rapid rate and it is alarming and vital to predict heart diseases early. The diagnosis of heart diseases is a challenging task i.e. it must be done accurately and proficiently. The aim of this study is to determine which patient is more likely to have heart disease based on a number of medical features. We organized a heart disease prediction model to identify whether the person is likely to be diagnosed with a (...) heart disease or not using the medical features of the person. We used many different algorithms of machine learning such as Gaussian Mixture, Nearest Centroid, MultinomialNB, Logistic RegressionCV, Linear SVC, Linear Discriminant Analysis, SGD Classifier, Extra Tree Classifier, Calibrated ClassifierCV, Quadratic Discriminant Analysis, GaussianNB, Random Forest Classifier, ComplementNB, MLP Classifier, BernoulliNB, Bagging Classifier, LGBM Classifier, Ada Boost Classifier, K Neighbors Classifier, Logistic Regression, Gradient Boosting Classifier, Decision Tree Classifier, and Deep Learning to predict and classify the patient with heart disease. A quite helpful approach was used to regulate how the model can be used to improve the accuracy of prediction of heart diseases in any person. The strength of the proposed model was very satisfying and was able to predict evidence of having a heart disease in a particular person by using Deep Learning and Random Forest Classifier which showed a good accuracy in comparison to the other used classifiers. The proposed heart disease prediction model will enhances medical care and reduces the cost. This study gives us significant knowledge that can help us predict the person with heart disease. The dataset was collected from Kaggle depository and the model is implemented using python. (shrink)

In finite probability theory, events are subsets S⊆U of the outcome set. Subsets can be represented by 1-dimensional column vectors. By extending the representation of events to two dimensional matrices, we can introduce "superposition events." Probabilities are introduced for classical events, superposition events, and their mixtures by using density matrices. Then probabilities for experiments or `measurements' of all these events can be determined in a manner exactly like in quantum mechanics (QM) using density matrices. Moreover the transformation of the density (...) matrices induced by the experiments or `measurements' is the Lüders mixture operation as in QM. And finally by moving the machinery into the n-dimensional vector space over ℤ₂, different basis sets become different outcome sets. That `non-commutative' extension of finite probability theory yields the pedagogical model of quantum mechanics over ℤ₂ that can model many characteristic non-classical results of QM. (shrink)

Plastic waste is a significant environmental problem for almost all countries; therefore, protecting the environment from the problem is crucial. The most sensible solution to these problems is substituting the natural aggregates with substantial plastic waste in various building materials. This study aimed to optimise the mixed design ratio of cement brick containing plastic waste as aggregate replacement. Plastic cement brick mixtures were prepared by the incorporation of four different types of plastic waste such as polyethylene terephthalate (PET), high-density polyethylene, (...) low-density polyethylene and polypropylene into cement bricks with different cement contents (150, 300 and 450 g) and plastic replacement percentages (0, 3 and 6%). Compressive strength and water absorption of the plastic cement bricks were analysed using a statistical model through the response surface methodology. It revealed the optimum cement brick mixed design is C3-1% PET with the compressive strength of 27.50 MPa and water absorption of 1.16%. The optimised plastic cement brick also satisfied the general ASTM C62-17 requirements for building bricks despite the higher porosity observed by the scanning electron microscopy. The results from Fourier transform infrared spectroscopy analysis also showed that the addition of the plastic waste into cement brick was unlikely to modify the chemical compound within the cement brick mixtures. Thus, the proposed mathematical model can predict the required hardened properties of plastic cement bricks and could lead to greater utilisation of plastic waste in building materials. (shrink)

The present study is aimed to scan the explanatory relevance of the concept of glocalization in some seminal works of George Ritzer. In the first instance, we will try to relate the manner in which Ritzer understands glocalization to the uses of other authors or other related concepts of the cultural globalization theory. On this occasion, we will reveal the cultural and philosophical assumptions, underlying Ritzer’s use of this concept: the understanding of the individual, mainly seen as a rational agent, (...) as well as the positive value attributed to the postmodern type of cultural mixture. We will further argue that, despite its intentions, the manner in which Ritzer defines glocalization is in fact very close to a homogenized conception of globalization. In addition, we will show that Ritzer eludes the explanatory dimension of glocalization, in favor of a descriptive stance, excessively used. We will give also a critical analysis of the way in which Ritzer attempts to enrich the explanatory quality of glocalization by linking it with a new concept that he elaborated, the grobalization. In the end, we would like to connect Ritzer’s concept of glocalization with a social/sociological model exposed by the French sociologist Alain Touraine, hoping to better clarify the mentioned problems. (shrink)

Coherence and correspondence are classical contenders as theories of truth. In this paper we examine them instead as interacting factors in the dynamics of belief across epistemic networks. We construct an agent-based model of network contact in which agents are characterized not in terms of single beliefs but in terms of internal belief suites. Individuals update elements of their belief suites on input from other agents in order both to maximize internal belief coherence and to incorporate ‘trickled in’ elements of (...) truth as correspondence. Results, though often intuitive, prove more complex than in simpler models (Hegselmann & Krause 2002, 2006; Grim, Singer, Reade & Fisher 2015). The optimistic finding is that pressures toward internal coherence can exploit and expand on small elements of truth as correspondence is introduced into epistemic networks. Less optimistic results show that pressures for coherence can also work directly against the incorporation of truth, particularly when coherence is established first and new data is introduced later. (shrink)

The goal of this paper is to defend the general tenet that time travelers cannot change the past within B-theoretical models of time, independently of how many temporal dimensions there are. Baron Pacific Philosophical Quarterly, 98, 129–147 offered a strong argument intended to reach this general conclusion. However, his argument does not cover a peculiar case, i.e. a B-theoretical one-dimensional model of time that allows for the presence of internal times. Loss Pacific Philosophical Quarterly, 96, 1–11 used the latter (...) model to argue that time travelers can change the past within such model. We show a way to debunk Loss’s argument, so that the general tenet about the impossibility of changing the past within B-theoretical models is maintained. (shrink)

In this article, I show that the semantics one adopts for mass terms constrains the metaphysical claims one can make about mixtures. I first expose why mixtures challenge a singularist approach based on mereological sums. After discussing an alternative, non-singularist approach, I take chemistry into account and explain how it changes our perspective on these issues.

Continued urbanization, along with anthropogenic global warming, has and will increase land surface temperature and air temperature anomalies in urban areas when compared to their rural surroundings, leading to Urban Heat Islands (UHI). UHI poses environmental and health risks, affecting both psychological and physiological aspects of human health. Thus, using a deep learning approach that considers morphological variables, this study predicts UHI intensity in 69 European cities from 2007 to 2021 and projects UHI impacts for 2050 and 2080. The research (...) employs Artificial Neural Networks, Deep Neural Networks, and Gated Recurrent Units, combining high-resolution 3D urban models with environmental data to analyze UHI trends. The results indicate strong associations between urban form, weather patterns, and UHI intensity, highlighting the need for customized urban planning and policy measures to reduce UHI impacts and foster sustainable urban settings. This research enhances understanding of UHI dynamics and serves as a valuable tool for urban planners and policymakers to address the challenges of climate change, urbanization, and air pollution, ultimately aiding in the improvement of health outcomes and building energy consumption. Moreover, the methodology effectively demonstrates the ability of the GRU to link its scores with UHI projections, offering crucial insights into potential health impacts. (shrink)

This paper analyses the phenomenology and epistemology of chatbots such as ChatGPT and Bard. The computational architecture underpinning these chatbots are large language models (LLMs), which are generative artificial intelligence (AI) systems trained on a massive dataset of text extracted from the Web. We conceptualise these LLMs as multifunctional computational cognitive artifacts, used for various cognitive tasks such as translating, summarizing, answering questions, information-seeking, and much more. Phenomenologically, LLMs can be experienced as a “quasi-other”; when that happens, users anthropomorphise (...) them. For most users, current LLMs are black boxes, i.e., for the most part, they lack data transparency and algorithmic transparency. They can, however, be phenomenologically and informationally transparent, in which case there is an interactional flow. Anthropomorphising and interactional flow can, in some users, create an attitude of (unwarranted) trust towards the output LLMs generate. We conclude this paper by drawing on the epistemology of trust and testimony to examine the epistemic implications of these dimensions. Whilst LLMs generally generate accurate responses, we observe two epistemic pitfalls. Ideally, users should be able to match the level of trust that they place in LLMs to the degree that LLMs are trustworthy. However, both their data and algorithmic opacity and their phenomenological and informational transparency can make it difficult for users to calibrate their trust correctly. The effects of these limitations are twofold: users may adopt unwarranted attitudes of trust towards the outputs of LLMs (which is particularly problematic when LLMs hallucinate), and the trustworthiness of LLMs may be undermined. (shrink)

Under what conditions does machine learning (ML) model opacity inhibit the possibility of explaining and understanding phenomena? In this article, I argue that nonepistemic values give shape to the ML opacity problem even if we keep researcher interests fixed. Treating ML models as an instance of doing model-based science to explain and understand phenomena reveals that there is (i) an external opacity problem, where the presence of inductive risk imposes higher standards on externally validating models, and (ii) an (...) internal opacity problem, where greater inductive risk demands a higher level of transparency regarding the inferences the model makes. (shrink)

In this conceptual paper, the traditional conceptualization of sustainable entrepreneurship is challenged because of a fundamental tension between processes involved in sustainable development and processes involved in entrepreneurship: the concept of sustainable business models contains a paradox, because sustainability involves the reduction of information asymmetries, whereas entrepreneurship involves enhanced and secured levels of information asymmetries. We therefore propose a new and integrated theory of sustainable entrepreneurship that overcomes this paradox. The basic argument is that environmental problems have to be (...) conceptualized as wicked problems or sustainability-related ecosystem failures. Because all actors involved in the entrepreneurial process are characterized by their epistemic insufficiency regarding the solving of these problems, the role of information in the sustainable entrepreneurial process changes. On the one hand, the reduction of information asymmetries primarily aims to enable actors to become critical of sustainable entrepreneurs’ actual business models. On the other hand, the epistemic insufficiency of sustainable entrepreneurs guarantees that information asymmetries remain as a source of new sustainable business opportunities. Three further characteristics of sustainable entrepreneurs are distinguished: sustainability and entrepreneurship-related risk-taking; sustainability and entrepreneurship-related self-efficacy; and the development of satisficing and open-ended solutions, together with multiple stakeholders. (shrink)

A key challenge for probabilistic causal models is to distinguish non-causal probabilistic dependencies from true causal relations. To accomplish this task, causal models are usually required to satisfy several constraints. Two prominent constraints are the causal Markov condition and the faithfulness condition. However, other constraints are also needed. One of these additional constraints is the causal sufficiency condition, which states that models must not omit any direct common causes of the variables they contain. In this paper, I (...) argue that the causal sufficiency condition is problematic: (1) it is incompatible with the requirement that the variables in a model must not stand in non-causal necessary dependence relations, such as mathematical or conceptual relations, or relations described in terms of supervenience or grounding, (2) it presupposes more causal knowledge as primitive than is actually needed to create adequate causal models, and (3) if models are only required to be causally sufficient, they cannot deal with cases where variables are probabilistically related by accident, such as Sober’s example of the relationship between bread prices in England and the sea level in Venice. I show that these problems can be avoided if causal models are required to be monotonic in the following sense: the causal relations occurring in a model M would not disappear if further variables were added to M. I give a definition of this monotonicity condition and conclude that causal models should be required to be monotonic rather than causally sufficient. (shrink)

This paper examines the potential educational uses of chat-based large language models (LLMs), moving past initial hype and skepticism. Although LLM outputs often evoke fascination and resemble human writing, they are unpredictable and must be used with discernment. Several metaphors—like calculators, cars, and drunk tutors—highlight distinct models for student interactions with LLMs, which we explore in the paper. We suggest that LLMs hold a potential in students’ learning by fostering proleptic reasoning through scaffolding, i.e., presenting a technological accompaniment (...) in anticipating and responding to potential objections to arguments. Here, the technical limitations of LLMs can be reframed as beneficial when fostering anticipatory reasoning. Whether their outputs are accurate or not, evaluating them stimulates learning. LLMs require students to critically engage, emphasizing analytical thinking over mere memorization. This interaction helps solidify knowledge. Additionally, we explore how engaging with LLMs can prepare students for constructive collective discussions and provide first steps in addressing epistemic injustices by highlighting potential research blind spots. Thus, while acknowledging the sociopolitical and ethical complexities of using LLMs in education, we suggest that when used in an informed way, they can promote critical thinking through anticipatory reasoning. (shrink)

Several philosophers of science claim that scientific toy models afford knowledge of possibility, but answers to the question of why toy models can be expected to competently play this role are scarce. The main line of reply is that toy models support possibility claims insofar as they are credible. I raise a challenge for this credibility-thesis, drawing on a familiar problem for imagination-based modal epistemologies, and argue that it remains unanswered in the current literature. The credibility-thesis has (...) a long way to go if it is to account for the epistemic merits of toy models. (shrink)

Analytical categories of scientific cultures have typically been used both exclusively and universally. For instance, when styles of scientific research are employed in attempts to understand and narrate science, styles alone are usually employed. This article is a thought experiment in interweaving categories. What would happen if rather than employ a single category, we instead investigated several categories simultaneously? What would we learn about the practices and theories, the agents and materials, and the political-technological impact of science if we analyzed (...) and applied styles, paradigms, and models simultaneously? I address these questions in general and for a specific case study: a brief history of systematics. (shrink)

It is suggested that the impetus to generate models is probably the most fundamental point of connection between mysticism and psychology. In their concern with the relation between ‘unseen’ realms and the ‘seen’, mystical maps parallel cognitive models of the relation between ‘unconscious’ and ‘conscious’ processes. The map or model constitutes an explanation employing terms current within the respective canon. The case of language mysticism is examined to illustrate the premise that cognitive models may benefit from an (...) understanding of the kinds of experiences gained, and explanatory concepts advanced, within mystical traditions. Language mysticism is of particular interest on account of the central role thought to be played by language in relation to self and the individual's construction of reality. The discussion focuses on traditions of language mysticism within Judaism, in which emphasis is placed on the deconstruction of language into primary elements and the overarching significance of the divine Name. Analysis of the detailed techniques used suggests ways in which multiple associations to any given word/concept were consciously explored in an altered state. It appears that these mystics were consciously engaging with what are normally preconscious cognitive processes, whereby schematic associations to sensory images or thoughts are activated. The testimony from their writings implies that these mystics experienced distortions of the sense of self , which may suggest that, in the normal state, ‘I’ is constructed in relation to the preconscious system of associations. Moreover, an important feature of Hebrew language mysticism is its emphasis on embodiment -- specific associations were deemed to exist between the letters and each structure of the body. Implications, first, for the relationship between language and self, and, second, for the role of embodiment in relation to self are discussed. The importance of the continual emphasis on the Name of God throughout the linguistic practices may have provided a means for effectively replacing the cognitive indexing function hypothesized here to be normally played by ‘I’ with a more transpersonal cognitive index, especially in relation to memory. (shrink)

The rapid rise of generative AI, including LLMs, has prompted a great deal of concern, both within and beyond academia. One of these concerns is that generative models embed, reproduce, and therein potentially perpetuate all manner of bias. The present study offers an alternative perspective: exploring the potential of LLMs to detect bias in human generated text. Our target is genetic essentialism in obesity discourse in Australian print media. We develop and deploy an LLM-based classification model to evaluate a (...) large sample of relevant articles (n=26,163). We show that our model detects genetic essentialist biases as reliably as human experts; and find that, while genes figure less prominently in popular discussions of obesity than previous work might suggest, when genetic information is invoked, it is often presented in a biased way. Implications for future work are discussed. (shrink)

This article takes off from Johan van Benthem’s ruminations on the interface between logic and cognitive science in his position paper “Logic and reasoning: Do the facts matter?”. When trying to answer Van Benthem’s question whether logic can be fruitfully combined with psychological experiments, this article focuses on a specific domain of reasoning, namely higher-order social cognition, including attributions such as “Bob knows that Alice knows that he wrote a novel under pseudonym”. For intelligent interaction, it is important that the (...) participants recursively model the mental states of other agents. Otherwise, an international negotiation may fail, even when it has potential for a win-win solution, and in a time-critical rescue mission, a software agent may depend on a teammate’s action that never materializes. First a survey is presented of past and current research on higher-order social cognition, from the various viewpoints of logic, artificial intelligence, and psychology. Do people actually reason about each other’s knowledge in the way proscribed by epistemic logic? And if not, how can logic and cognitive science productively work together to construct more realistic models of human reasoning about other minds? The paper ends with a delineation of possible avenues for future research, aiming to provide a better understanding of higher-order social reasoning. The methodology is based on a combination of experimental research, logic, computational cognitive models, and agent-based evolutionary models. (shrink)

In recent analytic literature on the Trinity we have seen a variety of "social" models of the Trinity. By contrast there are few "non-­‐social" models. One prominent "non-­‐social" view is Brian Leftow's "Latin Trinity." I argue that the name of Leftow's model is not sufficiently descriptive in light of diverse models within Latin speaking theology. Next, I develop a new "non-­‐social" model that is inspired by Richard of St. Victor's description of a person in conjunction with my (...) appropriating insights about indexicals from David Kaplan and John Perry. I point out that the copula in tokens of statements like, "I am the Father," is an ambiguous term and when used by a certain divine person a different proposition is affirmed. Central to this model is the claim that the copula bears the "is of identity" and the "is of numerical sameness without identity." Further, I show that Leftow's model employs two concepts of "person," a Lockean one and a Boethian one, and mine employs Richard of St. Victor's. I describe Leftow's model as a "hard non-­‐social" model and mine as a "soft non-­‐social" model that is nearer to some social models. I conclude that Leftow's model is not the lone candidate among "non-­‐social" models and that the variety of "non-­‐social" models has yet to be exhausted. (shrink)

Moralising accounts of depression include the idea that depression is a sin or the result of sin, and/or that it is the result of demonic possession which has occurred because of moral or spiritual failure. Increasingly some Christian communities, understandably concerned about the debilitating effects these views have on people with depression, have adopted secular folk psychiatry’s ‘medicalising’ campaign, emphasising that depression is an illness for which, like (so-called) physical illnesses, experients should not be held responsible. This paper argues that (...) both moralising and medicalising models of depression are intellectually and practically (pastorally and therapeutically) problematic, gesturing towards more promising emphases. (shrink)

When does opinion formation within an interacting group lead to consensus, polarization or fragmentation? The article investigates various models for the dynamics of continuous opinions by analytical methods as well as by computer simulations. Section 2 develops within a unified framework the classical model of consensus formation, the variant of this model due to Friedkin and Johnsen, a time-dependent version and a nonlinear version with bounded confidence of the agents. Section 3 presents for all these models major analytical (...) results. Section 4 gives an extensive exploration of the nonlinear model with bounded confidence by a series of computer simulations. An appendix supplies needed mathematical definitions, tools, and theorems. (shrink)

Political theory as a discipline has long been skeptical of computational methods. In this paper, I argue that it is time for theory to make a perspectival shift on these methods. Specifically, we should consider integrating recently developed generative large language models like GPT-4 as tools to support our creative work as theorists. Ultimately, I suggest that political theorists should embrace this technology as a method of supporting our capacity for creativity—but that we should do so in a way (...) that is mindful of the content and value of theorizing, the technical constraints of the models, and the ethical questions that the technology raises. (shrink)

A common view of how-possibly explanations in biology treats them as explanatorily incomplete. In addition to this interpretation of how-possibly explanation, I argue that there is another interpretation, one which features what I term “explanatory strategies.” This strategy-centered interpretation of how-possibly explanation centers on there being a different explanatory context within which how-possibly explanations are offered. I contend that, in conditions where this strategy context is recognized, how-possibly explanations can be understood as complete explanations. I defend this alternative interpretation by (...) analyzing the explanatory value of simple physical models the nineteenth century developmental biologist Wilhelm His constructed for animal development. (shrink)

According to our understanding of the everyday physical world, observable phenomena are underpinned by persistent objects that can be reidentified across time by observation of their distinctive properties. This understanding is reflected in classical mechanics, which posits that matter consists of persistent, reidentifiable particles. However, the mathematical symmetrization procedures used to describe identical particles within the quantum formalism have led to the widespread belief that identical quantum particles lack either persistence or reidentifiability. However, it has proved difficult to reconcile these (...) assertions with the fact that identical particles are routinely assumed to be reidentifiable in particular circumstances. For example, when two electrons move through a bubble chamber, each is said to generate a sequence of bubbles that is caused by one and the same particle. Moreover, neither of these assertions accounts for the mathematical form of the symmetrization procedures used to describe identical particles within the quantum framework, leaving open theoretical possibilities other than bosonic and fermionic behavior, such as paraparticles, which do not appear to be realized in nature. Here we propose the novel idea that both persistence and nonpersistence models must be employed in order to fully account for the behaviour of identical particles. Thus, identical particles are neither persistent nor nonpersistent. We prove the viability of this viewpoint by showing how Feynman's and Dirac's symmetrization procedures arise through a synthesis of a quantum treatment of these models, and by showing how reidentifiability emerges in a context-dependent manner. We further show that the persistence and nonpersistence models satisfy the key characteristics of Bohr's concept of complementarity, and thereby propose that the behavior of identical particles is a manifestation of a persistence-nonpersistence complementarity, analogous to Bohr's wave-particle complementarity for individual particles. Finally, we construct a precise parallel between these two complementarities, and detail their conceptual similarities and dissimilarities. (shrink)

The depth-bounded approach seeks to provide realistic models of reasoners. Recognizing that most useful logics are idealizations in that they are either undecidable or likely to be intractable, the approach accounts for how they can be approximated in practice by resource-bounded agents. The approach has been applied to Classical Propositional Logic (CPL), yielding a hierarchy of tractable depth-bounded approximations to that logic, which in turn has been based on a KE/KI system. -/- This Thesis shows that the approach can (...) be naturally extended to useful nonclassical logics such as First-Degree Entailment (FDE), the Logic of Paradox (LP), Strong Kleene Logic (K3 ) and Intuitionistic Propositional Logic (IPL). To do this, we introduce a KE/KI-style system for each of those logics such that: is formulated via signed formulae, consist of linear operational rules and branching structural rule(s), can be used as a direct-proof and a refutation method, and is interesting independently of the approach in that it has an exponential speed-up on its tableau system counterpart. The latter given that we introduce a new class of examples which we prove to be hard for all tableau systems sharing the V/& rules with the classical one, but easy for their analogous KE-style systems. Then we focus on showing that each of our KE/KI-style systems naturally yields a hierarchy of tractable depth-bounded approximations to the respective logic, in terms of the maximum number of allowed nested applications of the branching rule(s). The rule(s) express(es) a generalized rule of bivalence, is (are) essentially cut rule(s) and govern(s) the manipulation of virtual information, which is information that an agent does not hold but she temporarily assumes as if she held it. Intuitively, the more virtual information needs to be invoked via the branching rule(s), the harder the inference is for the agent. So, the nested application the branching rule(s) provides a sensible measure of inferential depth. We also show that each hierarchy approximating FDE, LP, and K3 , admits of a 5-valued non-deterministic semantics; whereas, paving the way for a semantical characterization of the hierarchy approximating IPL, we provide a 3-valued non-deterministic semantics for the full logic that fixes the meaning of the connectives without appealing to “structural” conditions. -/- Moreover, we show a super-polynomial lower bound for the strongest possible version of clausal tableaux on the well-known class of “truly fat” expressions (which are easy for KE), settling a problem left open in the literature. Further, we investigate a hierarchy of tractable depth-bounded approximations to CPL based only on KE. Finally, we propose a refinement of the p-simulation relation which is adequate to establish positive results about the superiority of a system over another with respect to proof-search. (shrink)

As the demand for explainable deep learning grows in the evaluation of language technologies, the value of a principled grounding for those explanations grows as well. Here we study the state-of-the-art in explanation for neural models for natural-language processing (NLP) tasks from the viewpoint of philosophy of science. We focus on recent evaluation work that finds brittleness in explanations obtained through attention mechanisms.We harness philosophical accounts of explanation to suggest broader conclusions from these studies. From this analysis, we assert (...) the impossibility of causal explanations from attention layers over text data. We then introduce NLP researchers to contemporary philosophy of science theories that allow robust yet non-causal reasoning in explanation, giving computer scientists a vocabulary for future research. (shrink)

I argue that Kant acknowledges two models of spontaneous self-determination that rational beings are capable of. The first model involves absolute unconditional necessity and excludes any form of contingency. The second model involves (albeit not as a matter of definition) a form of contingency which entails alternative possibilities for determining oneself. The first model would be exhibited by a divine being; the second model is exhibited by human beings. Human beings do, however, partake in the divine model up to (...) an extent, namely, as far as the legislative spontaneity of their law-giving faculties are concerned. I conclude by suggesting that for Kant, the mechanistic principle that we are exclusively determined by natural causes poses a twofold threat for human agency. In one respect (in relation to the second model), it threatens us with the obliteration of contingency, or with the universality of hypothetical necessity. But in another respect (in relation to the first model), it threatens us (and our putative "laws") with the obliteration of absolute necessity, or with the universality of contingency. (shrink)

How do people reason about their opponent in turn-taking games? Often, people do not make the decisions that game theory would prescribe. We present a logic that can play a key role in understanding how people make their decisions, by delineating all plausible reasoning strategies in a systematic manner. This in turn makes it possible to construct a corresponding set of computational models in a cognitive architecture. These models can be run and fitted to the participants’ data in (...) terms of decisions, response times, and answers to questions. We validate these claims on the basis of an earlier game-theoretic experiment about the turn-taking game “Marble Drop with Surprising Opponent”, in which the opponent often starts with a seemingly irrational move. We explore two ways of segregating the participants into reasonable “player types”. The first way is based on latent class analysis, which divides the players into three classes according to their first decisions in the game: Random players, Learners, and Expected players, who make decisions consistent with forward induction. The second way is based on participants’ answers to a question about their opponent, classified according to levels of theory of mind: zero-order, first-order and second-order. It turns out that increasing levels of decisions and theory of mind both correspond to increasing success as measured by monetary awards and increasing decision times. Next, we use the logical language to express different kinds of strategies that people apply when reasoning about their opponent and making decisions in turn-taking games, as well as the ‘reasoning types’ reflected in their behavior. Then, we translate the logical formulas into computational cognitive models in the PRIMs architecture. Finally, we run two of the resulting models, corresponding to the strategy of only being interested in one’s own payoff and to the myopic strategy, in which one can only look ahead to a limited number of nodes. It turns out that the participant data fit to the own-payoff strategy, not the myopic one. The article closes the circle from experiments via logic and cognitive modelling back to predictions about new experiments. (shrink)

According to the Lockean thesis, a proposition is believed just in case it is highly probable. While this thesis enjoys strong intuitive support, it is known to conflict with seemingly plausible logical constraints on our beliefs. One way out of this conflict is to make probability 1 a requirement for belief, but most have rejected this option for entailing what they see as an untenable skepticism. Recently, two new solutions to the conflict have been proposed that are alleged to be (...) non-skeptical. We compare these proposals with each other and with the Lockean thesis, in particular with regard to the question of how much we gain by adopting any one of them instead of the probability 1 requirement, that is, of how likely it is that one believes more than the things one is fully certain of. (shrink)