Joint Point Process Model for Counterfactual Treatment–Outcome Trajectories Under Policy Interventions

Caglar Hizli; Ti John; Anne Juuti; Tuure Saarinen; Kirsi Pietiläinen; Pekka Marttinen

Joint Point Process Model for Counterfactual Treatment–Outcome Trajectories Under Policy Interventions

Caglar Hizli^*, Ti John, Anne Juuti, Tuure Saarinen, Kirsi Pietiläinen, Pekka Marttinen

^*Tämän työn vastaava kirjoittaja

Tutkimustuotos: Artikkeli kirjassa/konferenssijulkaisussa › Konferenssiesitys › Scientific › vertaisarvioitu

Abstrakti

Policy makers need to predict the progression of an outcome before adopting
a new treatment policy, which defines when and how a sequence of treatments
affecting the outcome occurs in continuous time. Commonly, algorithms that
predict interventional future outcome trajectories take a fixed sequence of future treatments as input. This excludes scenarios where the policy is unknown or a counterfactual analysis is needed. To handle these limitations, we develop a joint model for treatments and outcomes, which allows for the estimation of treatment policies and effects from sequential treatment–outcome data. It can answer interventional and counterfactual queries about interventions on treatment policies, as we show with a realistic semi-synthetic simulation study. This abstract is based on work that is currently under review for AAAI-23.

Alkuperäiskieli	Englanti
Tila	Julkaistu - 2022
OKM-julkaisutyyppi	Ei sovellu
Tapahtuma	Conference on Neural Information Processing Systems - New Orleans, Yhdysvallat Kesto: 28 marrask. 2022 → 9 jouluk. 2022 Konferenssinumero: 36 https://nips.cc/

Conference

Conference	Conference on Neural Information Processing Systems
Lyhennettä	NeurIPS
Maa/Alue	Yhdysvallat
Kaupunki	New Orleans
Ajanjakso	28/11/2022 → 09/12/2022
www-osoite	https://nips.cc/

Pääsy asiakirjaan

https://gp-seminar-series.github.io/neurips-2022/assets/camera_ready/46.pdf

Siteeraa tätä

Hizli, C., John, T., Juuti, A., Saarinen, T., Pietiläinen, K., & Marttinen, P. (2022). Joint Point Process Model for Counterfactual Treatment–Outcome Trajectories Under Policy Interventions. Julkaisun esittämispaikka: Conference on Neural Information Processing Systems , New Orleans, Louisiana, Yhdysvallat. https://gp-seminar-series.github.io/neurips-2022/assets/camera_ready/46.pdf

@conference{bf9ffc86286f49089ceb82c75ad6fdc3,

title = "Joint Point Process Model for Counterfactual Treatment–Outcome Trajectories Under Policy Interventions",

abstract = "Policy makers need to predict the progression of an outcome before adoptinga new treatment policy, which defines when and how a sequence of treatmentsaffecting the outcome occurs in continuous time. Commonly, algorithms thatpredict interventional future outcome trajectories take a fixed sequence of future treatments as input. This excludes scenarios where the policy is unknown or a counterfactual analysis is needed. To handle these limitations, we develop a joint model for treatments and outcomes, which allows for the estimation of treatment policies and effects from sequential treatment–outcome data. It can answer interventional and counterfactual queries about interventions on treatment policies, as we show with a realistic semi-synthetic simulation study. This abstract is based on work that is currently under review for AAAI-23.",

author = "Caglar Hizli and Ti John and Anne Juuti and Tuure Saarinen and Kirsi Pietil{\"a}inen and Pekka Marttinen",

year = "2022",

language = "English",

note = "Conference on Neural Information Processing Systems , NeurIPS ; Conference date: 28-11-2022 Through 09-12-2022",

url = "https://nips.cc/",

}

Hizli, C , John, T, Juuti, A, Saarinen, T, Pietiläinen, K & Marttinen, P 2022, 'Joint Point Process Model for Counterfactual Treatment–Outcome Trajectories Under Policy Interventions', Artikkeli esitetty Conference on Neural Information Processing Systems , New Orleans, Yhdysvallat, 28/11/2022 - 09/12/2022. <https://gp-seminar-series.github.io/neurips-2022/assets/camera_ready/46.pdf>

Joint Point Process Model for Counterfactual Treatment–Outcome Trajectories Under Policy Interventions. / Hizli, Caglar ; John, Ti; Juuti, Anne et al.
2022. Julkaisun esittämispaikka: Conference on Neural Information Processing Systems , New Orleans, Louisiana, Yhdysvallat.

Tutkimustuotos: Artikkeli kirjassa/konferenssijulkaisussa › Konferenssiesitys › Scientific › vertaisarvioitu

TY - CONF

T1 - Joint Point Process Model for Counterfactual Treatment–Outcome Trajectories Under Policy Interventions

AU - Hizli, Caglar

AU - John, Ti

AU - Juuti, Anne

AU - Saarinen, Tuure

AU - Pietiläinen, Kirsi

AU - Marttinen, Pekka

N1 - Conference code: 36

PY - 2022

Y1 - 2022

N2 - Policy makers need to predict the progression of an outcome before adoptinga new treatment policy, which defines when and how a sequence of treatmentsaffecting the outcome occurs in continuous time. Commonly, algorithms thatpredict interventional future outcome trajectories take a fixed sequence of future treatments as input. This excludes scenarios where the policy is unknown or a counterfactual analysis is needed. To handle these limitations, we develop a joint model for treatments and outcomes, which allows for the estimation of treatment policies and effects from sequential treatment–outcome data. It can answer interventional and counterfactual queries about interventions on treatment policies, as we show with a realistic semi-synthetic simulation study. This abstract is based on work that is currently under review for AAAI-23.

AB - Policy makers need to predict the progression of an outcome before adoptinga new treatment policy, which defines when and how a sequence of treatmentsaffecting the outcome occurs in continuous time. Commonly, algorithms thatpredict interventional future outcome trajectories take a fixed sequence of future treatments as input. This excludes scenarios where the policy is unknown or a counterfactual analysis is needed. To handle these limitations, we develop a joint model for treatments and outcomes, which allows for the estimation of treatment policies and effects from sequential treatment–outcome data. It can answer interventional and counterfactual queries about interventions on treatment policies, as we show with a realistic semi-synthetic simulation study. This abstract is based on work that is currently under review for AAAI-23.

M3 - Paper

T2 - Conference on Neural Information Processing Systems

Y2 - 28 November 2022 through 9 December 2022

ER -

Joint Point Process Model for Counterfactual Treatment–Outcome Trajectories Under Policy Interventions

Abstrakti

Conference

Pääsy asiakirjaan

Sormenjälki

Siteeraa tätä