A state-based inverse reinforcement learning approach to model activity-travel choices behavior with reward function recovery

↓ Direkt till sidans innehåll
↓ Direkt till sidans sekundära innehåll (sidomenyn)

Sökning: onr:"swepub:oai:DiVA.org:kth-341608" > A state-based inver...

1 av 1
Föregående post
Nästa post
Till träfflistan

Song, YuchenKTH,Byggvetenskap,School of Transportation, Southeast University, Nanjing 211189, Jiangsu, China, Jiangsu (författare)

A state-based inverse reinforcement learning approach to model activity-travel choices behavior with reward function recovery

Artikel/kapitelEngelska2024

Förlag, utgivningsår, omfång ...

Elsevier BV,2024
printrdacarrier

Nummerbeteckningar

LIBRIS-ID:oai:DiVA.org:kth-341608
https://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-341608URI
https://doi.org/10.1016/j.trc.2023.104454DOI

Kompletterande språkuppgifter

Språk:engelska
Sammanfattning på:engelska

Ingår i deldatabas

SwePubSwePub

Klassifikation

Ämneskategori:ref swepub-contenttype
Ämneskategori:art swepub-publicationtype

Anmärkningar

QC 20231227
Behaviorally oriented activity-travel choices (ATC) modeling is a principal part of travel demand analysis. Traditional econometric and rule-based methods require explicit model structures and complex domain knowledge. While several recent studies used machine learning models, especially adversarial inverse reinforcement learning (IRL) models, to learn potential ATC patterns with less expert-designed settings, they lack a clear representation of rational ATC behavior. In this study, we propose a data-driven IRL framework based on the maximum causal approach to minimize f-divergences between expert and agent state marginal distributions, which provides a more sample-efficient measurement. In addition, we specify a separate state-only reward function and derive an analytical gradient of the f-divergence objective with respect to reward parameters to ensure good convergences. The method can recover a stationary reward function, which assures the agent to get close to the expert behavior when training from scratch. We validate the proposed model using cellular signaling data from Chongqing, China by comparing with baseline models (behavior cloning, policy-based, and reward-based models) in aspects of policy performance comparison, reward recovery, and reward transfer tasks. The experiment results indicate that the proposed model outperforms existing methods and is relatively less sensitive to the number of expert demonstrations. Qualitative analyses are provided on the fundamental ATC preferences on different features given the reward function recovered from the observed mobility trajectories, and on the learning behaviors under different choices of f-divergence.

Ämnesord och genrebeteckningar

TEKNIK OCH TEKNOLOGIER Samhällsbyggnadsteknik Transportteknik och logistik hsv//swe
ENGINEERING AND TECHNOLOGY Civil Engineering Transport Systems and Logistics hsv//eng
Activity-travel choices modeling
Cellular signaling data
Inverse reinforcement learning
Reward function
State marginal matching

Biuppslag (personer, institutioner, konferenser, titlar ...)

Li, DaweiSchool of Transportation, Southeast University, Nanjing 211189, Jiangsu, China, Jiangsu; Jiangsu Key Laboratory of Urban ITS, Southeast University, Nanjing 211189, Jiangsu, China, Jiangsu; Jiangsu Province Collaborative Innovation Center of Modern Urban Traffic Technologies, Nanjing 211189, Jiangsu, China, Jiangsu (författare)
Ma, ZhenliangKTH,Transportplanering(Swepub:kth)u12zc8sm (författare)
Liu, DongjieSchool of Transportation, Southeast University, Nanjing 211189, Jiangsu, China, Jiangsu (författare)
Zhang, TongSchool of Transportation, Southeast University, Nanjing 211189, Jiangsu, China, Jiangsu (författare)
KTHByggvetenskap (creator_code:org_t)

Sammanhörande titlar

Ingår i:Transportation Research Part C: Elsevier BV1580968-090X1879-2359

Internetlänk

Hitta via bibliotek

Transportation Research Part C (Sök värdpublikationen i LIBRIS)

Till lärosätets databas

1 av 1
Föregående post
Nästa post
Till träfflistan

Hitta mer i SwePub

Av författaren/redakt...: Song, Yuchen; Li, Dawei; Ma, Zhenliang; Liu, Dongjie; Zhang, Tong

Om ämnet

TEKNIK OCH TEKNOLOGIER: TEKNIK OCH TEKNO ...; och Samhällsbyggnads ...; och Transportteknik ...

Artiklar i publikationen: Transportation R ...

Av lärosätet: Kungliga Tekniska Högskolan

Sök utanför SwePub

Sök vidare i:: Google; Google Book Search; Google Scholar

Kungliga biblioteket hanterar dina personuppgifter i enlighet med EU:s dataskyddsförordning (2018), GDPR. Läs mer om hur det funkar här.
Så här hanterar KB dina uppgifter vid användning av denna tjänst.

LIBRIS.kb.se