Neural Interpretable Reasoning
This work addresses the challenge of interpretability in deep learning, which is crucial for building trust and transparency in AI systems, but it appears incremental as it builds on existing concepts like equivariance and Markovian properties.
The authors tackled the problem of achieving interpretability in deep learning by formalizing a framework based on inference equivariance, and they proposed a new modeling paradigm called neural generation and interpretable execution that enables scalable verification of equivariance.
We formalize a novel modeling framework for achieving interpretability in deep learning, anchored in the principle of inference equivariance. While the direct verification of interpretability scales exponentially with the number of variables of the system, we show that this complexity can be mitigated by treating interpretability as a Markovian property and employing neural re-parametrization techniques. Building on these insights, we propose a new modeling paradigm -- neural generation and interpretable execution -- that enables scalable verification of equivariance. This paradigm provides a general approach for designing Neural Interpretable Reasoners that are not only expressive but also transparent.