Targeted Causal Elicitation

Nazaal Ibrahim; Ti John; Zhigao Guo; Samuel Kaski

Targeted Causal Elicitation

Nazaal Ibrahim, Ti John, Zhigao Guo, Samuel Kaski

Machine Learning and Robotics

Aalto University

Research output: Contribution to conference › Paper › peer-review

Abstract

We look at the problem of learning causal structure for a fixed downstream causal effect optimization task. In contrast to previous work which often focuses on running interventional experiments, we consider an often overlooked source of information - the domain expert. In the Bayesian setting, this amounts to augmenting the likelihood with a user model whose parameters account for possible biases of the expert. Such a model can allow for active elicitation in a manner that is most informative to the optimization task at hand.

Original language	English
Number of pages	8
Publication status	Accepted/In press - 2022
Event	NeurIPS Workshop on Causality for Real-world Impact - Duration: 2 Dec 2022 → 2 Dec 2022

Conference

Conference	NeurIPS Workshop on Causality for Real-world Impact
Period	2/12/22 → 2/12/22

Keywords

Machine learning

Access to Document

https://openreview.net/pdf?id=oqRw-a4rf36

Turing AI Fellowship: Human-AI Research Teams - Steering AI in Experimental Design and Decision-Making
Kaski, S. (PI), Bristow, R. (CoI), Cai, P. (CoI), Jay, C. (CoI) & Peek, N. (CoI)
1/10/21 → 30/09/26
Project: Research

Cite this

@conference{30d474089eb642b383a91161aee5602d,

title = "Targeted Causal Elicitation",

abstract = "We look at the problem of learning causal structure for a fixed downstream causal effect optimization task. In contrast to previous work which often focuses on running interventional experiments, we consider an often overlooked source of information - the domain expert. In the Bayesian setting, this amounts to augmenting the likelihood with a user model whose parameters account for possible biases of the expert. Such a model can allow for active elicitation in a manner that is most informative to the optimization task at hand.",

keywords = "Machine learning",

author = "Nazaal Ibrahim and Ti John and Zhigao Guo and Samuel Kaski",

year = "2022",

language = "English",

note = "NeurIPS Workshop on Causality for Real-world Impact ; Conference date: 02-12-2022 Through 02-12-2022",

}

TY - CONF

T1 - Targeted Causal Elicitation

AU - Ibrahim, Nazaal

AU - John, Ti

AU - Guo, Zhigao

AU - Kaski, Samuel

PY - 2022

Y1 - 2022

N2 - We look at the problem of learning causal structure for a fixed downstream causal effect optimization task. In contrast to previous work which often focuses on running interventional experiments, we consider an often overlooked source of information - the domain expert. In the Bayesian setting, this amounts to augmenting the likelihood with a user model whose parameters account for possible biases of the expert. Such a model can allow for active elicitation in a manner that is most informative to the optimization task at hand.

AB - We look at the problem of learning causal structure for a fixed downstream causal effect optimization task. In contrast to previous work which often focuses on running interventional experiments, we consider an often overlooked source of information - the domain expert. In the Bayesian setting, this amounts to augmenting the likelihood with a user model whose parameters account for possible biases of the expert. Such a model can allow for active elicitation in a manner that is most informative to the optimization task at hand.

KW - Machine learning

M3 - Paper

T2 - NeurIPS Workshop on Causality for Real-world Impact

Y2 - 2 December 2022 through 2 December 2022

ER -

Targeted Causal Elicitation

Abstract

Conference

Keywords

Access to Document

Fingerprint

Projects

Turing AI Fellowship: Human-AI Research Teams - Steering AI in Experimental Design and Decision-Making

Cite this