Triadic Concept Analysis for Logic Interpretation of Simple Artificial Networks
This addresses the lack of interpretability in ANNs for users needing transparent decision-making, though it is incremental as it builds on existing formal concept analysis methods.
The paper tackled the problem of interpretability in artificial neural networks by deriving symbolic representations from simple ANN models, resulting in logic trees that preserve the classification accuracy of the original models.
An artificial neural network (ANN) is a numerical method used to solve complex classification problems. Due to its high classification power, the ANN method often outperforms other classification methods in terms of accuracy. However, an ANN model lacks interpretability compared to methods that use the symbolic paradigm. Our idea is to derive a symbolic representation from a simple ANN model trained on minterm values of input objects. Based on ReLU nodes, the ANN model is partitioned into cells. We convert the ANN model into a cell-based, three-dimensional bit tensor. The theory of Formal Concept Analysis applied to the tensor yields concepts that are represented as logic trees, expressing interpretable attribute interactions. Their evaluations preserve the classification power of the initial ANN model.