Abstract

Invariant representation learning (IRL) encourages the prediction from invariant causal features to labels deconfounded from the environments, advancing the technical roadmap of out-of-distribution (OOD) generalization. Despite spotlights around, recent theoretical result verified that some causal features recovered by IRLs merely pretend domain-invariantly in the training environments but fail in unseen domains. The fake invariance severely endangers OOD generalization since the trustful objective can not be diagnosed and existing causal remedies are invalid to rectify. In this paper, we review a IRL family (InvRat) under the Partially and Fully Informative Invariant Feature Structural Causal Models (PIIF SCM /FIIF SCM) respectively, to certify their weaknesses in representing fake invariant features, then, unify their causal diagrams to propose ReStructured SCM (RS-SCM). RS-SCM can ideally rebuild the spurious and the fake invariant features simultaneously. Given this, we further develop an approach based on conditional mutual information with respect to RS-SCM, then rigorously rectify the spurious and fake invariant effects. It can be easily implemented by a small feature selection subnet introduced in the IRL family, which is alternatively optimized to achieve our goal. Experiments verified the superiority of our approach to fight against the fake invariant issue across a variety of OOD generalization benchmarks.

Framework

Experiment

Conclusion

To prick flter bubbles in the CRS, we propose a novel framework FacetCRS, which models multi-faceted user preferences in the CRS, including entity-, word-, context-, and review-facet, to capture diverse user preferences to mitigate flter bubbles. Meanwhile, FacetCRS is an end-to-end framework to automatically learn representations of various levels of preference facet and diverse types of external knowledge. Through extensive experiments, our method consistently outperforms several competitive base\lines, which demonstrate the effectiveness of our FacetCRS.

Acknowledgement

This work is supported in part by the National Key R&D Program of China under Grant No. 2021ZD0111601, National Natural Science Foundation of China (NSFC) under Grant No. 61836012, 62325605, U21A20470, 62206110, 62206314, GuangDong Basic and Applied Basic Research Foundation under Grant No. 2023A1515011374, 2022A1515011835, China Postdoctoral Science Foundation funded project under Grant No. 2021M703687.