Extracting Formulae in Many-Valued Logic from Deep Neural Networks
Not on the map. Why?
Abstract (arXiv)
We propose a new perspective on deep ReLU networks, namely as circuit counterparts of Lukasiewicz infinite-valued logic -- a many-valued (MV) generalization of Boolean logic. An algorithm for extracting formulae in MV logic from deep ReLU networks is presented. As the algorithm applies to networks with general, in particular also real-valued, weights, it can be used to extract logical formulae from deep ReLU networks trained on data.
Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 14 of its 50 references that are in the map. First posted on arXiv 22 Jan 2024.
Builds on
14 of 50 references.
References · 14
- 2023Sparks of Artificial General Intelligence: Early experiments with GPT-4cited by 2,592
- 2016Error bounds for approximations with deep ReLU networkscited by 574
- 2017Nonparametric regression using deep neural networks with ReLU activation functioncited by 433
- 2016On the Expressive Power of Deep Neural Networkscited by 306
- 2019Deep Learning for Symbolic Mathematicscited by 270
- 2022Transformers Learn Shortcuts to Automatacited by 268
- 2021Neural Symbolic Regression that Scalescited by 192
- 2021Thinking Like Transformerscited by 184
- 2020Logical Neural Networkscited by 119
- 2016Why does deep and cheap learning work so well?cited by 100
- 2019Deep Neural Network Approximation Theorycited by 93
- 2015Representation Benefits of Deep Feedforward Networkscited by 86
Built on by
Citing papers per year: 2026: 1.
1 paper.