The Risks and Rewards of Invariant Risk Minimization
Spurious correlations are one of the most prominent pain points for building and deploying machine learning models. Invariant Risk Minimization (IRM) is a learning algorithm designed to mitigate the effect of spurious features and perform well despite shifts in the test distribution. In this blog post, we discuss the motivation and details of IRM as well as it's criticisms and shortcomings.
Code (0)
등록된 구현이 없습니다.
Tasks
BIG-bench Machine LearningSimilar Papers 제목 키워드 기반
The Risks of Invariant Risk Minimization
Invariant Causal Prediction (Peters et al., 2016) is a technique for out-of-distribution generalization which assumes that some aspects of the data distribution vary across the training set but that the underlying causal…
Out-of-Distribution GeneralizationInvariant Information Bottleneck for Domain Generalization
Invariant risk minimization (IRM) has recently emerged as a promising alternative for domain generalization. Nevertheless, the loss function is difficult to optimize for nonlinear classifiers and the original optimizatio…
Domain GeneralizationContinual Invariant Risk Minimization
Empirical risk minimization can lead to poor generalization behavior on unseen environments if the learned model does not capture invariant feature representations. Invariant risk minimization (IRM) is a recent proposal …
Continual LearningAsymptotic Analysis of Risk Premia Induced by Law-Invariant Risk Measures
We analyze the limiting behavior of the risk premium associated with the Pareto optimal risk sharing contract in an infinitely expanding pool of risks under a general class of law-invariant risk measures encompassing ran…
Kernelized Heterogeneous Risk Minimization
The ability to generalize under distributional shifts is essential to reliable machine learning, while models optimized with empirical risk minimization usually fail on non-$i.i.d$ testing data. Recently, invariant learn…