Learning to Defer for Causal Discovery with Imperfect Experts

• URL: http://arxiv.org/abs/2502.13132v1
• Date: Tue, 18 Feb 2025 18:55:53 GMT
• Title: Learning to Defer for Causal Discovery with Imperfect Experts
• Authors: Oscar Clivio, Divyat Mahajan, Perouz Taslakian, Sara Magliacane, Ioannis Mitliagkas, Valentina Zantedeschi, Alexandre Drouin,
• Abstract summary: We propose L2D-CD, a method for gauging the correctness of expert recommendations and optimally combining them with data-driven causal discovery results. We evaluate L2D-CD on the canonical T"ubingen pairs dataset and demonstrate its superior performance compared to both the causal discovery method and the expert used in isolation.
• Score: 59.071731337922664
• License:
• Abstract: Integrating expert knowledge, e.g. from large language models, into causal discovery algorithms can be challenging when the knowledge is not guaranteed to be correct. Expert recommendations may contradict data-driven results, and their reliability can vary significantly depending on the domain or specific query. Existing methods based on soft constraints or inconsistencies in predicted causal relationships fail to account for these variations in expertise. To remedy this, we propose L2D-CD, a method for gauging the correctness of expert recommendations and optimally combining them with data-driven causal discovery results. By adapting learning-to-defer (L2D) algorithms for pairwise causal discovery (CD), we learn a deferral function that selects whether to rely on classical causal discovery methods using numerical data or expert recommendations based on textual meta-data. We evaluate L2D-CD on the canonical T\"ubingen pairs dataset and demonstrate its superior performance compared to both the causal discovery method and the expert used in isolation. Moreover, our approach identifies domains where the expert's performance is strong or weak. Finally, we outline a strategy for generalizing this approach to causal discovery on graphs with more than two variables, paving the way for further research in this area.

Related papers

• Discovery of Maximally Consistent Causal Orders with Large Language Models [0.8192907805418583]
  Causal discovery is essential for understanding complex systems. Traditional methods often rely on strong, untestable assumptions. We propose a novel method to derive a class of acyclic tournaments.
  arXiv  Detail & Related papers  (2024-12-18T16:37:51Z)
• Sample, estimate, aggregate: A recipe for causal discovery foundation models [28.116832159265964]
  We train a supervised model that learns to predict a larger causal graph from the outputs of classical causal discovery algorithms run over subsets of variables. Our approach is enabled by the observation that typical errors in the outputs of classical methods remain comparable across datasets. Experiments on real and synthetic data demonstrate that this model maintains high accuracy in the face of misspecification or distribution shift.
  arXiv  Detail & Related papers  (2024-02-02T21:57:58Z)
• Adversarial Imitation Learning On Aggregated Data [0.0]
  Inverse Reinforcement Learning (IRL) learns an optimal policy, given some expert demonstrations, thus avoiding the need for the tedious process of specifying a suitable reward function. We propose an approach which removes these requirements through a dynamic, adaptive method called Adversarial Imitation Learning on Aggregated Data (AILAD) It learns conjointly both a non linear reward function and the associated optimal policy using an adversarial framework.
  arXiv  Detail & Related papers  (2023-11-14T22:13:38Z)
• Human-in-the-Loop Causal Discovery under Latent Confounding using Ancestral GFlowNets [15.95243318673688]
  Most causal discovery algorithms do not provide uncertainty estimates, making it hard for users to interpret results and improve the inference process. We propose to sample (causal) ancestral graphs proportionally to a belief distribution based on a score function, such as the Bayesian information criterion (BIC) We then introduce an optimal experimental design to iteratively probe the expert about the relations among variables, effectively reducing the uncertainty of our belief over ancestral graphs.
  arXiv  Detail & Related papers  (2023-09-21T12:53:45Z)
• Causal Discovery with Language Models as Imperfect Experts [119.22928856942292]
  We consider how expert knowledge can be used to improve the data-driven identification of causal graphs. We propose strategies for amending such expert knowledge based on consistency properties. We report a case study, on real data, where a large language model is used as an imperfect expert.
  arXiv  Detail & Related papers  (2023-07-05T16:01:38Z)
• Improving Few-Shot Generalization by Exploring and Exploiting Auxiliary Data [100.33096338195723]
  We focus on Few-shot Learning with Auxiliary Data (FLAD) FLAD assumes access to auxiliary data during few-shot learning in hopes of improving generalization. We propose two algorithms -- EXP3-FLAD and UCB1-FLAD -- and compare them with prior FLAD methods that either explore or exploit.
  arXiv  Detail & Related papers  (2023-02-01T18:59:36Z)
• Valid Inference After Causal Discovery [73.87055989355737]
  We develop tools for valid post-causal-discovery inference. We show that a naive combination of causal discovery and subsequent inference algorithms leads to highly inflated miscoverage rates.
  arXiv  Detail & Related papers  (2022-08-11T17:40:45Z)
• Principled Knowledge Extrapolation with GANs [92.62635018136476]
  We study counterfactual synthesis from a new perspective of knowledge extrapolation. We show that an adversarial game with a closed-form discriminator can be used to address the knowledge extrapolation problem. Our method enjoys both elegant theoretical guarantees and superior performance in many scenarios.
  arXiv  Detail & Related papers  (2022-05-21T08:39:42Z)
• Leveraging Expert Consistency to Improve Algorithmic Decision Support [62.61153549123407]
  We explore the use of historical expert decisions as a rich source of information that can be combined with observed outcomes to narrow the construct gap. We propose an influence function-based methodology to estimate expert consistency indirectly when each case in the data is assessed by a single expert. Our empirical evaluation, using simulations in a clinical setting and real-world data from the child welfare domain, indicates that the proposed approach successfully narrows the construct gap.
  arXiv  Detail & Related papers  (2021-01-24T05:40:29Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.