skip to main content


Title: KCRL: A Prior Knowledge Based Causal Discovery Framework With Reinforcement Learning
Causal discovery is an important problem in many sciences that enables us to estimate causal relationships from observational data. Particularly, in the healthcare domain, it can guide practitioners in making informed clinical decisions. Several causal discovery approaches have been developed over the last few decades. The success of these approaches mostly relies on a large number of data samples. In practice, however, an infinite amount of data is never available. Fortunately, often we have some prior knowledge available from the problem domain. Particularly, in healthcare settings, we often have some prior knowledge such as expert opinions, prior RCTs, literature evidence, and systematic reviews about the clinical problem. This prior information can be utilized in a systematic way to address the data scarcity problem. However, most of the existing causal discovery approaches lack a systematic way to incorporate prior knowledge during the search process. Recent advances in reinforcement learning techniques can be explored to use prior knowledge as constraints by penalizing the agent for their violations. Therefore, in this work, we propose a framework KCRL that utilizes the existing knowledge as a constraint to penalize the search process during causal discovery. This utilization of existing information during causal discovery reduces the graph search space and enables a faster convergence to the optimal causal mechanism. We evaluated our framework on benchmark synthetic and real datasets as well as on a real-life healthcare application. We also compared its performance with several baseline causal discovery methods. The experimental findings show that penalizing the search process for constraint violation yields better performance compared to existing approaches that do not include prior knowledge.  more » « less
Award ID(s):
2118285
NSF-PAR ID:
10343976
Author(s) / Creator(s):
;
Date Published:
Journal Name:
Proceedings of Machine Learning Research
Volume:
182
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. Fairness-aware machine learning has attracted a surge of attention in many domains, such as online advertising, personalized recommendation, and social media analysis in web applications. Fairness-aware machine learning aims to eliminate biases of learning models against certain subgroups described by certain protected (sensitive) attributes such as race, gender, and age. Among many existing fairness notions, counterfactual fairness is a popular notion defined from a causal perspective. It measures the fairness of a predictor by comparing the prediction of each individual in the original world and that in the counterfactual worlds in which the value of the sensitive attribute is modified. A prerequisite for existing methods to achieve counterfactual fairness is the prior human knowledge of the causal model for the data. However, in real-world scenarios, the underlying causal model is often unknown, and acquiring such human knowledge could be very difficult. In these scenarios, it is risky to directly trust the causal models obtained from information sources with unknown reliability and even causal discovery methods, as incorrect causal models can consequently bring biases to the predictor and lead to unfair predictions. In this work, we address the problem of counterfactually fair prediction from observational data without given causal models by proposing a novel framework CLAIRE. Specifically, under certain general assumptions, CLAIRE effectively mitigates the biases from the sensitive attribute with a representation learning framework based on counterfactual data augmentation and an invariant penalty. Experiments conducted on both synthetic and real-world datasets validate the superiority of CLAIRE in both counterfactual fairness and prediction performance. 
    more » « less
  2. Causal inference is at the heart of empirical research in natu- ral and social sciences and is critical for scientific discovery and informed decision making. The gold standard in causal inference is performing randomized controlled trials; unfortu- nately these are not always feasible due to ethical, legal, or cost constraints. As an alternative, methodologies for causal inference from observational data have been developed in sta- tistical studies and social sciences. However, existing meth- ods critically rely on restrictive assumptions such as the study population consisting of homogeneous elements that can be represented in a single flat table, where each row is referred to as a unit. In contrast, in many real-world set- tings, the study domain naturally consists of heterogeneous elements with complex relational structure, where the data is naturally represented in multiple related tables. In this paper, we present a formal framework for causal inference from such relational data. We propose a declarative language called CaRL for capturing causal background knowledge and assumptions, and specifying causal queries using simple Datalog-like rules. CaRL provides a foundation for infer- ring causality and reasoning about the effect of complex interventions in relational domains. We present an extensive experimental evaluation on real relational data to illustrate the applicability of CaRL in social sciences and healthcare. 
    more » « less
  3. Accurate pose estimation is often a requirement for robust robotic grasping and manipulation of objects placed in cluttered, tight environments, such as a shelf with multiple objects. When deep learning approaches are employed to perform this task, they typically require a large amount of training data. However, obtaining precise 6 degrees of freedom for ground-truth can be prohibitively expensive. This work therefore proposes an architecture and a training process to solve this issue. More precisely, we present a weak object detector that enables localizing objects and estimating their 6D poses in cluttered and occluded scenes. To minimize the human labor required for annotations, the proposed detector is trained with a combination of synthetic and a few weakly annotated real images (as little as 10 images per object), for which a human provides only a list of objects present in each image (no time-consuming annotations, such as bounding boxes, segmentation masks and object poses). To close the gap between real and synthetic images, we use multiple domain classifiers trained adversarially. During the inference phase, the resulting class-specific heatmaps of the weak detector are used to guide the search of 6D poses of objects. Our proposed approach is evaluated on several publicly available datasets for pose estimation. We also evaluated our model on classification and localization in unsupervised and semi-supervised settings. The results clearly indicate that this approach could provide an efficient way toward fully automating the training process of computer vision models used in robotics. 
    more » « less
  4. Background. While educational change often involves bold talk about disruptive ideas that eventually need to be institutionalized, a critical but often less visible element of sustaining change is work such as maintaining a shared vision, onboarding new people, negotiating small issues in light of department culture, and coordinating big changes with existing efforts. While knowledge about these forms of invisible work exist in other disciplines, these issues seem understudied in engineering education. This work approaches this issue of invisible knowledge with a design orientation, and specifically draws on the field of design-based research. Increasingly, design is recognized as a knowledge producing activity, resulting in insights into generative ways of defining problems, frameworks for generating solutions to problems, examples of what it looks like to connect theory to specific problems. Purpose: As a design effort, this work asks: How might a specific department create a sustainable practice to support the invisible work of coordinating and sustaining change? As a scholarly effort, this instance of design can result in a culminating problem definition, a solution framework, and examples of theory use that represent knowledge contributions. Approach: A mechanical engineering department in a small, private educational institution worked for four months to develop a sustainable practice to support invisible work of coordinating and sustaining change. Following an initial commitment of 60 minutes once every three weeks and 3-hour retreat to explore possibilities, the department then iteratively designed and then carried out sample conversations. Each iteration involved specifying the goals of the conversation, how to have the conversation (the design) and the rationale for connecting the design to the goals. Traces from the process represent the data for this work. Results. Over time, the conversations came to be designed along four dimensions: topic, time allocation, turn-taking, and traces. We have learned that topics that are of immediate relevance to everyone are particularly powerful (initial topics included "being back on campus" and "navigating in-person"). We are currently leveraging a time allocation that devotes the most time to hearing from each participant on the topic, then time for the group to cautiously explore synthesis, and finally time for the group to weigh in on future conversation topics. Approaches to turn-taking have involved decentralization (e.g., each current speaker invites the next speaker) and respect (speakers have a chance to "pass" and then choose the next speaker). Finally, we are experimenting with how to balance the creation of traces as a natural part of the process, such as through real-time transcription in the chat feature of zoom. Undergirding each of these dimensions are connections to the intended goals, connections to relevant theory, and connections to the long-term goal of sustainability. In presenting these ideas, we will focus on how the information being offered connects to the current body of knowledge in engineering education. Conclusion. It is promising to treat the work of department culture as a design problem. The ideas in this framework may serve as inspiration to others seeking to create their own sustainable mechanisms but with different conditions. During the winter and spring of 2022, the approach will be additionally tested via six deployments, and insights will be shared in subsequent publications. 
    more » « less
  5. Background. While educational change often involves bold talk about disruptive ideas that eventually need to be institutionalized, a critical but often less visible element of sustaining change is work such as maintaining a shared vision, onboarding new people, negotiating small issues in light of department culture, and coordinating big changes with existing efforts. While knowledge about these forms of invisible work exist in other disciplines, these issues seem understudied in engineering education. This work approaches this issue of invisible knowledge with a design orientation, and specifically draws on the field of design-based research. Increasingly, design is recognized as a knowledge producing activity, resulting in insights into generative ways of defining problems, frameworks for generating solutions to problems, examples of what it looks like to connect theory to specific problems. Purpose: As a design effort, this work asks: How might a specific department create a sustainable practice to support the invisible work of coordinating and sustaining change? As a scholarly effort, this instance of design can result in a culminating problem definition, a solution framework, and examples of theory use that represent knowledge contributions. Approach: A mechanical engineering department in a small, private educational institution worked for four months to develop a sustainable practice to support invisible work of coordinating and sustaining change. Following an initial commitment of 60 minutes once every three weeks and 3-hour retreat to explore possibilities, the department then iteratively designed and then carried out sample conversations. Each iteration involved specifying the goals of the conversation, how to have the conversation (the design) and the rationale for connecting the design to the goals. Traces from the process represent the data for this work. Results. Over time, the conversations came to be designed along four dimensions: topic, time allocation, turn-taking, and traces. We have learned that topics that are of immediate relevance to everyone are particularly powerful (initial topics included "being back on campus" and "navigating in-person"). We are currently leveraging a time allocation that devotes the most time to hearing from each participant on the topic, then time for the group to cautiously explore synthesis, and finally time for the group to weigh in on future conversation topics. Approaches to turn-taking have involved decentralization (e.g., each current speaker invites the next speaker) and respect (speakers have a chance to "pass" and then choose the next speaker). Finally, we are experimenting with how to balance the creation of traces as a natural part of the process, such as through real-time transcription in the chat feature of zoom. Undergirding each of these dimensions are connections to the intended goals, connections to relevant theory, and connections to the long-term goal of sustainability. In presenting these ideas, we will focus on how the information being offered connects to the current body of knowledge in engineering education. Conclusion. It is promising to treat the work of department culture as a design problem. The ideas in this framework may serve as inspiration to others seeking to create their own sustainable mechanisms but with different conditions. During the winter and spring of 2022, the approach will be additionally tested via six deployments, and insights will be shared in subsequent publications. 
    more » « less