On causal and anticausal learning

B. Schölkopf; D. Janzing; Jonas Martin Peters; E. Sgouritsa; K. Zhang; J.M. Mooij

On causal and anticausal learning

B. Schölkopf, D. Janzing, Jonas Martin Peters, E. Sgouritsa, K. Zhang, J.M. Mooij

88 Citations (Scopus)

Abstract

We consider the problem of function estimation in the case where an underlying causal model can be inferred. This has implications for popular scenarios such as covariate shift, concept drift, transfer learning and semi-supervised learning. We argue that causal knowledge may facilitate some approaches for a given problem, and rule out others. In particular, we formulate a hypothesis for when semi-supervised learning can help, and corroborate it with empirical results.

Original language	Undefined/Unknown
Title of host publication	Proceedings of the 29th International Conference on Machine Learning (ICML)
Number of pages	8
Publication date	2012
Pages	1255-1262
Publication status	Published - 2012
Externally published	Yes

Cite this

@inbook{67eccd2156024e0b9354a1b3b1b167b4,

title = "On causal and anticausal learning",

abstract = "We consider the problem of function estimation in the case where an underlying causal model can be inferred. This has implications for popular scenarios such as covariate shift, concept drift, transfer learning and semi-supervised learning. We argue that causal knowledge may facilitate some approaches for a given problem, and rule out others. In particular, we formulate a hypothesis for when semi-supervised learning can help, and corroborate it with empirical results.",

author = "B. Sch{\"o}lkopf and D. Janzing and Peters, {Jonas Martin} and E. Sgouritsa and K. Zhang and J.M. Mooij",

year = "2012",

language = "Udefineret/Ukendt",

pages = "1255--1262",

booktitle = "Proceedings of the 29th International Conference on Machine Learning (ICML)",

}

TY - CHAP

T1 - On causal and anticausal learning

AU - Schölkopf, B.

AU - Janzing, D.

AU - Peters, Jonas Martin

AU - Sgouritsa, E.

AU - Zhang, K.

AU - Mooij, J.M.

PY - 2012

Y1 - 2012

N2 - We consider the problem of function estimation in the case where an underlying causal model can be inferred. This has implications for popular scenarios such as covariate shift, concept drift, transfer learning and semi-supervised learning. We argue that causal knowledge may facilitate some approaches for a given problem, and rule out others. In particular, we formulate a hypothesis for when semi-supervised learning can help, and corroborate it with empirical results.

AB - We consider the problem of function estimation in the case where an underlying causal model can be inferred. This has implications for popular scenarios such as covariate shift, concept drift, transfer learning and semi-supervised learning. We argue that causal knowledge may facilitate some approaches for a given problem, and rule out others. In particular, we formulate a hypothesis for when semi-supervised learning can help, and corroborate it with empirical results.

M3 - Bidrag til bog/antologi

SP - 1255

EP - 1262

BT - Proceedings of the 29th International Conference on Machine Learning (ICML)

ER -