What KAN mortality say: smooth and interpretable mortality modeling using Kolmogorov-Arnold networks

<?xml version="1.0" encoding="UTF-8"?><modsCollection xmlns="http://www.loc.gov/mods/v3" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.loc.gov/mods/v3 http://www.loc.gov/standards/mods/v3/mods-3-8.xsd">
<mods version="3.8">
<titleInfo>
<title>What KAN mortality say: smooth and interpretable mortality modeling using Kolmogorov-Arnold networks</title>
</titleInfo>
<name type="personal" xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="MAPA20260001678">
<namePart>Zhuang, Yuan</namePart>
<nameIdentifier>MAPA20260001678</nameIdentifier>
</name>
<name type="corporate" xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="MAPA20100017661">
<namePart>International Actuarial Association</namePart>
<nameIdentifier>MAPA20100017661</nameIdentifier>
</name>
<typeOfResource>text</typeOfResource>
<genre authority="marcgt">periodical</genre>
<originInfo>
<place>
<placeTerm type="code" authority="marccountry">bel</placeTerm>
</place>
<dateIssued encoding="marc">2026</dateIssued>
<issuance>serial</issuance>
</originInfo>
<language>
<languageTerm type="code" authority="iso639-2b">eng</languageTerm>
</language>
<physicalDescription>
<form authority="marcform">print</form>
</physicalDescription>
<abstract displayLabel="Summary">This study explores how traditional mortality model components can be directly used to initialize neural networks, addressing the common loss of structural information that occurs during model interpretation. We introduce KolmogorovArnold Networks (KAN) and develop both shallow models (KAN[2,1] and ARIMAKAN) and a KAN-based Actuarial Neural Network (KANN), which extends the existing CANN framework. The proposed KANN variantsKANN[2,1], KANNLC, and KANNAPCincorporate elements from classical mortality models and enhance them through deep learning. These models produce smooth mortality curves and smooth age, period, and cohort effects through simple regularization. Experiments across 34 populations show that KAN-based approaches offer stable performance, successfully balancing interpretability, smoothness, and predictive accuracy</abstract>
<note type="statement of responsibility">Lianzeng Zhang and Yuan Zhuang</note>
<subject xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="MAPA20080555306">
<topic>Mortalidad</topic>
</subject>
<subject xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="MAPA20080592042">
<topic>Modelos matemáticos</topic>
</subject>
<subject xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="MAPA20080579258">
<topic>Cálculo actuarial</topic>
</subject>
<subject xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="MAPA20170005476">
<topic>Machine learning</topic>
</subject>
<subject xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="MAPA20080555016">
<topic>Longevidad</topic>
</subject>
<subject xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="MAPA20080602437">
<topic>Matemática del seguro</topic>
</subject>
<subject xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="MAPA20080597733">
<topic>Modelos estadísticos</topic>
</subject>
<classification authority="">6</classification>
<relatedItem type="host">
<titleInfo>
<title>Astin bulletin</title>
</titleInfo>
<originInfo>
<publisher>Belgium : ASTIN and AFIR Sections of the International Actuarial Association</publisher>
</originInfo>
<identifier type="issn">0515-0361</identifier>
<identifier type="local">MAP20077000420</identifier>
<part>
<text>19/01/2026 Volume 56 Issue 1 - January 2026 , p. 32 - 59</text>
</part>
</relatedItem>
<recordInfo>
<recordContentSource authority="marcorg">MAP</recordContentSource>
<recordCreationDate encoding="marc">260202</recordCreationDate>
<recordChangeDate encoding="iso8601">20260205101745.0</recordChangeDate>
<recordIdentifier source="MAP">MAP20260002064</recordIdentifier>
<languageOfCataloging>
<languageTerm type="code" authority="iso639-2b">spa</languageTerm>
</languageOfCataloging>
</recordInfo>
</mods>
</modsCollection>