arXiv subjects

Jun Otsuka

Publications and source records attributed to Jun Otsuka.

8 recordsLinked to original sources

arXiv2024

What Machine Learning Tells Us About the Mathematical Structure of Concepts

This paper examines the connections among various approaches to understanding concepts in philosophy, cognitive science, and machine learning, with a particular focus on their mathematical nature. By categorizing these approaches into Abstractionism, the Similarity Approach, the Functional Approach, and the Invariance Approach, the study highlights how each framework provides a distinct mathematical perspective for modeling concepts. The synthesis of these approaches bridges philosophical theories and contemporary machine learning models, providing a comprehensive framework for future research. This work emphasizes the importance of interdisciplinary dialogue, aiming to enrich our understanding of the complex relationship between human cognition and artificial intelligence.

cs.AI↗

arXiv2024

Collective Predictive Coding as Model of Science: Formalizing Scientific Activities Towards Generative Science

This paper proposes a new conceptual framework called Collective Predictive Coding as a Model of Science (CPC-MS) to formalize and understand scientific activities. Building on the idea of collective predictive coding originally developed to explain symbol emergence, CPC-MS models science as a decentralized Bayesian inference process carried out by a community of agents. The framework describes how individual scientists' partial observations and internal representations are integrated through communication and peer review to produce shared external scientific knowledge. Key aspects of scientific practice like experimentation, hypothesis formation, theory development, and paradigm shifts are mapped onto components of the probabilistic graphical model. This paper discusses how CPC-MS provides insights into issues like social objectivity in science, scientific progress, and the potential impacts of AI on research. The generative view of science offers a unified way to analyze scientific activities and could inform efforts to automate aspects of the scientific process. Overall, CPC-MS aims to provide an intuitive yet formal model of science as a collective cognitive activity.

physics.soc-ph↗

arXiv2022

On the Equivalence of Causal Models: A Category-Theoretic Approach

We develop a category-theoretic criterion for determining the equivalence of causal models having different but homomorphic directed acyclic graphs over discrete variables. Following Jacobs et al. (2019), we define a causal model as a probabilistic interpretation of a causal string diagram, i.e., a functor from the ``syntactic'' category $\textsf{Syn}_G$ of graph $G$ to the category $\textsf{Stoch}$ of finite sets and stochastic matrices. The equivalence of causal models is then defined in terms of a natural transformation or isomorphism between two such functors, which we call a $Φ$-abstraction and $Φ$-equivalence, respectively. It is shown that when one model is a $Φ$-abstraction of another, the intervention calculus of the former can be consistently translated into that of the latter. We also identify the condition under which a model accommodates a $Φ$-abstraction, when transformations are deterministic.

cs.LG↗

arXiv2020

Rational Choice Hypothesis as X-point of Utility Function and Norm Function

Towards the realization of a sustainable, fair and inclusive society, we proposed a novel decision-making model that incorporates social norms in a rational choice model from the standpoints of deontology and utilitarianism. We proposed a hypothesis that interprets choice of action as the X-point for individual utility function that increases with actions and social norm function that decreases with actions. This hypothesis is based on humans psychologically balancing the value of utility and norms in selecting actions. Using the hypothesis and approximation, we were able to isolate and infer utility function and norm function from real-world measurement data of actions on environmental conditions and elucidate the interaction between the both functions that led from current status to target actions. As examples of collective data that aggregate decision-making of individuals, we looked at the changes in power usage before and after the Great East Japan Earthquake and the correlation between national GDP and CO2 emission in different countries. The first example showed that the perceived benefits of power (i.e., utility of power usage) was stronger than the power usage restrictions imposed by norms after the earthquake, contrary to our expectation. The second example showed that a reduction of CO2 emission in each country was not related to utility derived from GDP but to norms related to CO2 emission. Going forward, we will apply this new X-point model to actual social practices involving normative problems, and design the approaches for the diagnosis, prognosis and intervention of social systems by IT systems.

econ.GN↗

arXiv2020

Sustainability and Fairness Simulations Based on Decision-Making Model of Utility Function and Norm Function

We introduced a decision-making model based on value functions that included individualistic utility function and socio-constructivistic norm function and proposed a norm-fostering process that recursively updates norm function through mutual recognition between the self and others. As an example, we looked at the resource-sharing problem typical of economic activities and assumed the distribution of individual actions to define the (1) norm function fostered through mutual comparison of value/action ratio based on the equity theory (progressive tax-like), (2) norm function proportional to resource utilization (proportional tax-like) and (3) fixed norm function independent of resource utilization (fixed tax-like). By carrying out numerical simulation, we showed that the progressive tax-like norm function (i) does not increase disparity for the distribution of the actions, unlike the other norm functions, and (ii) has high resource productivity and low Gini coefficient. Therefore the progressive tax-like norm function has the highest sustainability and fairness.

econ.GN↗

arXiv2017

Multi-level computational methods for interdisciplinary research in the HathiTrust Digital Library

We show how faceted search using a combination of traditional classification systems and mixed-membership topic models can go beyond keyword search to inform resource discovery, hypothesis formulation, and argument extraction for interdisciplinary research. Our test domain is the history and philosophy of scientific work on animal mind and cognition. The methods can be generalized to other research areas and ultimately support a system for semi-automatic identification of argument structures. We provide a case study for the application of the methods to the problem of identifying and extracting arguments about anthropomorphism during a critical period in the development of comparative psychology. We show how a combination of classification systems and mixed-membership models trained over large digital libraries can inform resource discovery in this domain. Through a novel approach of "drill-down" topic modeling---simultaneously reducing both the size of the corpus and the unit of analysis---we are able to reduce a large collection of fulltext volumes to a much smaller set of pages within six focal volumes containing arguments of interest to historians and philosophers of comparative psychology. The volumes identified in this way did not appear among the first ten results of the keyword search in the HathiTrust digital library and the pages bear the kind of "close reading" needed to generate original interpretations that is the heart of scholarly work in the humanities. Zooming back out, we provide a way to place the books onto a map of science originally constructed from very different data and for different purposes. The multilevel approach advances understanding of the intellectual and societal contexts in which writings are interpreted.

cs.DL↗

arXiv2008

First-principles calculation of electronic polarization of III-V nanotubes

A first-principles study of the electronic polarization of BN and AlN nanotubes and their graphitic sheets under an external electric field has been performed. We found that the polarization per atom of zigzag nanotubes increases with decreasing diameter while that of armchair nanotubes decreases. The variation of the polarization is related to the exterior angle of the bonds around the B or Al atoms rather than that around the N atoms. The increase in the polarization of the zigzag nanotubes with decreasing diameter is caused by the large variation of the exterior angle when they are wrapped into the tubular form. On the other hand, the decrease in the bond length results in the weak polarization of thin armchair nanotubes.

cond-mat.mtrl-sci↗

arXiv2004

First-principles study on field evaporation of surface atoms from W(011) and Mo(011) surfaces

The simulations of field-evaporation processes for surface atoms on W(011) and Mo(011) surfaces are implemented using first-principles calculations based on the real-space finite-difference method. The threshold values of the external electric field for evaporation of the surface atoms, which are $\sim$ 6 V/Å2mm for tungsten and $\sim$ 5 V/Å2mm for molybdenum, are in agreement with the experimental results. Whereas field evaporation has been believed to occur as a result of significant local-field enhancement around the evaporating atoms, in this study, the enhancement is not observed around the atoms but above them and the strength of the local field is much smaller than that expected on the basis of the classical model.

cond-mat.mtrl-sci↗