Understanding Task Representations in Neural Networks via Bayesian Ablation
Neural networks are powerful tools for cognitive modeling due to their flexibility and emergent properties. However, interpreting their learned representations remains challenging due to their sub-symbolic semantics. In this work, we introduce a novel probabilistic framework for interpreting latent task representations in neural networks. Inspired by Bayesian inference, our approach defines a distribution over representational units to infer their causal contributions to task performance. Using ideas from information theory, we propose a suite of tools and metrics to illuminate key model properties, including representational distributedness, manifold complexity, and polysemanticity.
Code (0)
등록된 구현이 없습니다.
Tasks
Bayesian InferenceSimilar Papers 제목 키워드 기반
A Bit More Bayesian: Domain-Invariant Learning with Uncertainty
Domain generalization is challenging due to the domain shift and the uncertainty caused by the inaccessibility of target domain data. In this paper, we address both challenges with a probabilistic framework based on vari…
Bayesian InferenceDomain GeneralizationDeep Bayesian Network for Visual Question Generation
Generating natural questions from an image is a semantic task that requires using vision and language modalities to learn multimodal representations. Images can have multiple visual and language cues such as places, capt…
Natural QuestionsQuestion GenerationQuestion-GenerationAsymptotics of representation learning in finite Bayesian neural networks
Recent works have suggested that finite Bayesian neural networks may sometimes outperform their infinite cousins because finite networks can flexibly adapt their internal representations. However, our theoretical underst…
Representation LearningEvent Temporal Relation Extraction with Bayesian Translational Model
Existing models to extract temporal relations between events lack a principled method to incorporate external knowledge. In this study, we introduce Bayesian-Trans, a Bayesian learning-based method that models the tempor…
Bayesian InferencemodelRelationRelation Extraction+2Ablation Studies for Novel Treatment Effect Estimation Models
Ablation studies are essential for understanding the contribution of individual components within complex models, yet their application in nonparametric treatment effect estimation remains limited. This paper emphasizes …
Uncertainty Quantification