---
res:
  bibo_abstract:
  - Causal representation learning (CRL) aims at identifying high-level causal variables
    from low-level data, e.g. images. Current methods usually assume that all causal
    variables are captured in the high-dimensional observations. In this work, we
    focus on learning causal representations from data under partial observability,
    i.e., when some of the causal variables are not observed in the measurements,
    and the set of masked variables changes across the different samples. We introduce
    some initial theoretical results for identifying causal variables under partial
    observability by exploiting a sparsity regularizer, focusing in particular on
    the linear and piecewise linear mixing function case. We provide a theorem that
    allows us to identify the causal variables up to permutation and element-wise
    linear transformations in the linear case and a lemma that allows us to identify
    causal variables up to linear transformation in the piecewise case. Finally, we
    provide a conjecture that would allow us to identify the causal variables up to
    permutation and element-wise linear transformations also in the piecewise linear
    case. We test the theorem and conjecture on simulated data, showing the effectiveness
    of our method.@eng
  bibo_authorlist:
  - foaf_Person:
      foaf_givenName: Danru
      foaf_name: Xu, Danru
      foaf_surname: Xu
  - foaf_Person:
      foaf_givenName: Dingling
      foaf_name: Yao, Dingling
      foaf_surname: Yao
      foaf_workInfoHomepage: http://www.librecat.org/personId=d3e02e50-48a8-11ee-8f62-c108061797fa
  - foaf_Person:
      foaf_givenName: Sebastien
      foaf_name: Lachapelle, Sebastien
      foaf_surname: Lachapelle
  - foaf_Person:
      foaf_givenName: Perouz
      foaf_name: Taslakian, Perouz
      foaf_surname: Taslakian
  - foaf_Person:
      foaf_givenName: Julius
      foaf_name: von Kügelgen, Julius
      foaf_surname: von Kügelgen
  - foaf_Person:
      foaf_givenName: Francesco
      foaf_name: Locatello, Francesco
      foaf_surname: Locatello
      foaf_workInfoHomepage: http://www.librecat.org/personId=26cfd52f-2483-11ee-8040-88983bcc06d4
    orcid: 0000-0002-4850-0683
  - foaf_Person:
      foaf_givenName: Sara
      foaf_name: Magliacane, Sara
      foaf_surname: Magliacane
  dct_date: 2023^xs_gYear
  dct_language: eng
  dct_publisher: OpenReview@
  dct_title: A sparsity principle for partially observable causal representation learning@
...