Research map / Architectures and efficiency / RNNs and normalization
Inverse Abstraction of Neural Networks Using Symbolic Interpolation
Sumanth Dathathri, Sicun Gao, Richard M. Murray
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| increase nodes per hidden layer (5 to 20) | abstraction computation · 2D toy | run-time | Hurt |
| increase number of hidden layers (2 to 4) | abstraction computation · 2D toy | run-time | Hurt |
| replace NN controller with controller from overapproximation of {left} | cart-pole controller · Cart-pole | reward: 200.0 -> 200.0 | No clear effect |
| verify robustness with both under- and overapproximations | abstractions · 2D toy | points decided: 49 of 50 points decided | Improved |
| underapproximation threshold 0.3 instead of 0.5 on theta_dot | cart-pole abstractions · Cart-pole | property check: counterexample with -0.31 | Hurt |
Closest papers by meaning
- Network Inversion of Convolutional Neural Nets (Student Abstract) (AAAI 2025 · similarity 0.84)
- Explaining, Fast and Slow: Abstraction and Refinement of Provable Explanations (ICML 2025 · similarity 0.83)
- Taming Binarized Neural Networks and Mixed-Integer Programs (AAAI 2024 · similarity 0.83)
- Transformers Converge to Invariant Algorithmic Cores (NeurIPS 2026 · similarity 0.82)
- OrthoReg: Orthogonal Regularization for Hybrid Symbolic-Neural Dynamical Systems (NeurIPS 2026 · similarity 0.82)
- SMILE: Bridging Continuous Optimization and Discrete Symbolic Recovery (NeurIPS 2026 · similarity 0.82)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
