8 - Seminar Meta Learning (SemMeL) - Lena Eichermüller - A Meta-Transfer Objective for Learning to Disentangle Causal Mechanisms/ClipID:27253 previous clip next clip

Keywords: meta learning
Recording date 2020-12-25

Language

English

Organisational Unit

Friedrich-Alexander-Universität Erlangen-Nürnberg

Producer

Friedrich-Alexander-Universität Erlangen-Nürnberg

Abstract

We propose to meta-learn causal structures based on how fast a learner adapts to new distributions arising from sparse distributional changes, e.g. due to interventions, actions of agents and other sources of non-stationarities. We show that under this assumption, the correct causal structural choices lead to faster adaptation to modified distributions because the changes are concentrated in one or just a few mechanisms when the learned knowledge is modularized appropriately. This leads to sparse expected gradients and a lower effective number of degrees of freedom needing to be relearned while adapting to the change. It motivates using the speed of adaptation to a modified distribution as a meta-learning objective. We demonstrate how this can be used to determine the cause-effect relationship between two observed variables. The distributional changes do not need to correspond to standard interventions (clamping a variable), and the learner has no direct knowledge of these interventions. We show that causal structures can be parameterized via continuous variables and learned end-to-end. We then explore how these ideas could be used to also learn an encoder that would map low-level observed variables to unobserved causal variables leading to faster adaptation out-of-distribution, learning a representation space where one can satisfy the assumptions of independent mechanisms and of small and sparse changes in these mechanisms due to actions and non-stationarities.

https://arxiv.org/abs/1901.10912

More clips in this category "Friedrich-Alexander-Universität Erlangen-Nürnberg"

2024-02-07
IdM-login
protected  
2024-02-06
IdM-login / Studon
protected  
2024-02-06
Studon
protected  
2024-02-06
Passwort / Studon
protected