Symbolic knowledge extraction from trained neural networks: A sound approach
From MaRDI portal
Publication:5940782
DOI10.1016/S0004-3702(00)00077-1zbMath0969.68124OpenAlexW2065204741WikidataQ102362850 ScholiaQ102362850MaRDI QIDQ5940782
Krysia Broda, Artur S. d'Avila Garcez, Dov M. Gabbay
Publication date: 20 August 2001
Published in: Artificial Intelligence (Search for Journal in Brave)
Full work available at URL: https://doi.org/10.1016/s0004-3702(00)00077-1
Related Items
Learning from interpretation transition using differentiable logic programming semantics ⋮ Logic programs and connectionist networks ⋮ Logic programs, iterated function systems, and recurrent radial basis function networks ⋮ Is it worth generating rules from neural network ensembles? ⋮ Connectionist modal logic: representing modalities in neural networks ⋮ Extracting and inserting knowledge into stacked denoising auto-encoders ⋮ Neurosymbolic integration with uncertainty ⋮ Symbolic knowledge extraction from trained neural networks: A sound approach ⋮ Connectionist computations of intuitionistic reasoning ⋮ DEFINITION AND ADAPTATION OF WEIGHTED FUZZY LOGIC PROGRAMS ⋮ Expressing Belief Flow in Assertion Networks ⋮ A Connectionist Computational Model for Epistemic and Temporal Reasoning ⋮ Unnamed Item ⋮ Piecewise-Linear Neural Networks and Their Relationship to Rule Extraction from Data ⋮ Comprehensibility maximization and humanly comprehensible representations
Cites Work
- Knowledge-based artificial neural networks
- Multilayer feedforward networks are universal approximators
- Horn approximations of empirical data
- A Penalty-Function Approach for Pruning Feedforward Neural Networks
- Extracting Rules from Neural Networks by Pruning and Hidden-Unit Splitting
- Argument-based extended logic programming with defeasible priorities
- Symbolic knowledge extraction from trained neural networks: A sound approach
- Unnamed Item
- Unnamed Item
- Unnamed Item
- Unnamed Item
- Unnamed Item
- Unnamed Item
