Causal Learning for Socially Responsible AI
arXiv:2104.12278
Abstract
There have been increasing concerns about Artificial Intelligence (AI) due to its unfathomable potential power. To make AI address ethical challenges and shun undesirable outcomes, researchers proposed to develop socially responsible AI (SRAI). One of these approaches is causal learning (CL). We survey state-of-the-art methods of CL for SRAI. We begin by examining the seven CL tools to enhance the social responsibility of AI, then review how existing works have succeeded using these tools to tackle issues in developing SRAI such as fairness. The goal of this survey is to bring forefront the potentials and promises of CL for SRAI.
8 pages, 3 figures, accepted at IJCAI21 survey track
References in corpus (16)
- Counterfactual Fairness
- Invariant Risk Minimization
- Out-of-Distribution Generalization via Risk Extrapolation (REx)
- An introduction to domain adaptation and transfer learning
- A Survey of Learning Causality with Data: Problems and Methods
- Recommendations as Treatments: Debiasing Learning and Evaluation
- The Risks of Invariant Risk Minimization
- The Deconfounded Recommender: A Causal Inference Approach to Recommendation
- Causality Learning: A New Perspective for Interpretable Machine Learning
- Survey on Causal-based Machine Learning Fairness Notions
- Socially Responsible AI Algorithms: Issues, Purposes, and Challenges
- Generative Counterfactual Introspection for Explainable Deep Learning
- Counterfactual Generative Networks
- Explaining Deep Learning Models using Causal Inference
- An Empirical Study of Invariant Risk Minimization
- Sample-Efficient Reinforcement Learning via Counterfactual-Based Data Augmentation