Robustness to Spurious Correlations via Human Annotations
By Megha Srivastava, Tatsunori B. Hashimoto, Percy Liang. 2020
103 citations recorded by the source corpus.
The reliability of machine learning systems critically assumes that the associations between features and labels remain similar between training and test distributions. However, unmeasured variables, such as confounders, break this assumption---useful correlations between features and labels at training time can become useless or even harmful at test time. For example, high obesity is generally predictive for heart disease, but this relation may not hold for smokers who generally have lower rates of obesity and higher rates of heart disease. We present a framework for making models robust to spurious correlations by leveraging humans' common sense knowledge of causality. Specifically, we use human annotation to augment each training example with a potential unmeasured variable (i.e. an underweight patient with heart disease may be a smoker), reducing the problem to a covariate shift problem. We then introduce a new distributionally robust optimization objective over unmeasured variables (UV-DRO) to control the worst-case loss over possible test-time shifts. Empirically, we show improvements of 5-10% on a digit recognition task confounded by rotation, and 1.5-5% on the task of analyzing NYPD Police Stops confounded by location.
Subfields: Artificial Intelligence, Safety, Robustness & Interpretability
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Percy Liang
- Nicholas Carlini
- Mohit Bansal
- Quoc V. Le
- Ludwig Schmidt
- Eduard H. Hovy
- Richard S. Zemel
- Yoshua Bengio
In Safety, Robustness & Interpretability
- Nicholas Carlini
- Jörn-Henrik Jacobsen
- Richard S. Zemel
- Ludwig Schmidt
- Vaishaal Shankar
- B. Recht
- Achal Dave
- Elliot Creager
Citations in the corpus
- Measuring Robustness to Natural Distribution Shifts in Image Classification (2020, methodology): “[81] Srivastava, M.”
- Distilling Model Failures as Directions in Latent Space (2022, unknown): “Moreover, identifying meaningful subpopulations within data allows for dataset refinement (such as filtering or relabeling) (Yang et al., 2019; Stock & Cisse, 2018), and training more fair (Kim et al., 2019; Du et al., 2021) or accurate (Jabbour et al., 2020; Srivastava et al., 2020) models.”
- Environment Inference for Invariant Learning (2020, background): “In some cases, relevant side-information or metadata, e.g., human annotations, or device ID used to take a medical image, hospital or department ID, etc., may be abundant, but it remains unclear how best to specify environments based on this information (Srivastava et al., 2020).”
- Explaining The Efficacy of Counterfactually-Augmented Data (2020, background): “One recent line of work (Kaushik et al., 2020; Teney et al., 2020; Srivastava et al., 2020) has sought to inject causal thinking into real world settings by leveraging human-in-the-loop feedback to identify those signals that cause a label to be applicable versus those that merely happen to be…”