SAM: Structural Agnostic Model, Causal Discovery and Penalized Adversarial Learning

By Diviyan Kalainathan, Olivier Goudet, Isabelle M Guyon, David Lopez-Paz, Michèle Sebag. 2018

47 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Causal Inference & Discovery

Citations in the corpus

Differentiable Causal Discovery from Interventional Data (2020, methodology): “The work of [44] was limited to linear causal relationships, but was quickly extended to nonlinear ones via neural networks [21, 41, 45, 24, 18, 46].”
Gradient-Based Neural DAG Learning (2019, background): “SAM (Kalainathan et al., 2018) learns conditional NN generators using adversarial losses but does not enforce acyclicity.”
Learning Neural Causal Models from Unknown Interventions (2019, background): “…based on observational data (Chickering, 2002; Tsamardinos et al., 2006; Goudet et al., 2017; Hauser & Bühlmann, 2012; Spirtes et al., 2000; Sun et al., 2007; Zhang et al., 2012; Shimizu et al., 2006; Hoyer et al., 2009; Daniusis et al., 2012; Budhathoki & Vreeken, 2017; Kalainathan et al., 2018).”
DAG-GNN: DAG Structure Learning with Graph Neural Networks (2019, methodology): “Kalainathan et al. (2018) propose a GAN-style (generative adversarial network) method, whereby a separate generative model is applied to each variable and a discrimi-nator is used to distinguish between the joint distributions of real and generated samples.”
Learning Sparse Nonparametric DAGs (2019, methodology): “Our approach is based on a recent algebraic characterization of DAGs that led to a fully continuous program for scorebased learning of DAG models parametrized by a linear structural equation model (SEM).”
Causal Discovery Toolbox: Uncover causal relationships in Python (2019, unknown): “, such as CGNN or SAM (Goudet et al., 2017; Kalainathan et al., 2018).”
Causal Discovery with Attention-Based Convolutional Neural Networks (2019, methodology): “Although called ‘causal filters’ by the authors, SAM uses an attention mechanism by multiplying each observed input variable by a trainable score, comparable to the TCDF approach.”
Automated Machine Learning: Methods, Systems, Challenges (2019, unknown): “In stark contrast to this attitude, there is a recent tendency (apart from the above link to thermodynamics) to consider causality as crucial for a better understanding of physics: in exploring the foundations of quantum theory, researchers have adopted framework and ideas from causal inference [23]”
Causal Induction from Visual Observations for Goal Directed Tasks (2019, unknown): “In addition, adversarial learning (Kalainathan et al., 2018), graph neural networks (Yu et al., 2019), and gradient-based DAG learning (Lachapelle et al., 2019) have been recently introduced to causal discovery, but largely focusing on small synthetic datasets.”