Abstract We propose quantitative probing as a model-agnostic framework for validating causal models in the presence of domain knowledge. The method is constructed analogy to train/test split correlation-based machine learning. It consistent with logic scientific discovery and enhances current validation strategies. effectiveness illustrated using Pearl’s sprinkler example, before thorough simul...

Large capacity machine learning (ML) models are prone to membership inference attacks (MIAs), which aim infer whether the target sample is a member of model's training dataset. The serious privacy concerns due have motivated multiple defenses against MIAs, e.g., differential and adversarial regularization. Unfortunately, these produce ML with unacceptably low classification performances. Our wo...

With the further development of knowledge graphs, many weighted graphs (WKGs) have been published and greatly promote various applications. However, current deterministic graph embedding algorithms cannot encode well. This paper gives a promising framework WeExt that can extend models to enable them learn embeddings. In addtion, we introduce link prediction evaluate models’ performance on compl...