Intrinsic interpretability of graph neural networks (GNNs) is to find a small subset of the input graph's features -rationale -which guides the model prediction. Unfortunately, the leading rationalization models often rely on data biases, especially shortcut features, to compose rationales and make predictions without probing the critical and causal patterns. Moreover, such data biases easily change outside the training distribution. As a result, these models suffer from a huge drop in interpretability and predictive performance on out-of-distribution data. In this work, we propose a new strategy of discovering invariant rationale (DIR) to construct intrinsically interpretable GNNs. It conducts interventions on the training distribution to create multiple interventional distributions. Then it approaches the causal rationales that are invariant across different distributions while filtering out the spurious patterns that are unstable. Experiments on both synthetic and realworld datasets validate the superiority of our DIR in terms of interpretability and generalization ability on graph classification over the leading baselines. Code and datasets are available at https://github.com/Wuyxin/DIR-GNN.
Antioxidant responses induced by decabromodiphenyl ether (BDE-209) in the earthworms (Eisenia fetida) were studied after 7 days of exposure. Electron paramagnetic resonance (EPR) spectra indicated that hydroxyl radicals (•OH) in earthworms were significantly induced by 0.01-10 mg/kg of BDE-209. Malondialdehyde (MDA) and protein carbonyl (PCO) were stimulated at 0.5-10 mg/kg and 1-10 mg/kg, respectively. The reduced glutathione (GSH) was inhibited at 1-10 mg/kg while oxidized glutathione (GSSG) accumulated at 0.5-10 mg/kg. The GSH/GSSG ratio decreased at 0.5-10 mg/kg, and superoxide dismutase (SOD) and glutathione-S-transferase (GST) activities were induced at 0.1-1 and 0.5 mg/kg, respectively and both decreased at 10 mg/kg. Catalase (CAT) activities increased at 1-10 mg/kg. The results showed that severe oxidative stress occurred in E. fetida, and may play an important role in inducing the toxicity of BDE-209 on earthworms.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.