SymbolNet: neural symbolic regression with adaptive dynamic pruning for compression
Problem
Finding compact symbolic expressions for high-dimensional datasets is hard for genetic-programming-based symbolic regression, which matters for efficient inference on resource-constrained hardware such as FPGAs at the CERN Large Hadron Collider.
Why it matters
Efficient, accurate models on constrained hardware enable real-time data processing in high-energy physics experiments.
What the AI did
The paper proposes SymbolNet, a neural-network approach to symbolic regression that scales to high-dimensional inputs and uses gradient methods with adaptive dynamic pruning to constrain expression complexity.
Results
not stated in source (abstract truncated before concrete results)
Impact
Evidence supports a scalable AI method for compact expression discovery, but measured results are cut off.
Record
| Venue | Machine Learning Science and Technology |
|---|---|
| Published | 2025 |
| Work type | article |
| Authors | Ho Fung Tsoi, Vladimir Lončar, S. Dasu, Philip Coleman Harris |
| Citations | 13 (OpenAlex, at import time) |
| References | 90 |
| Model or method | SymbolNet (neural network-based symbolic regression) |
| Organization | not stated in source |
| Primary source | https://doi.org/10.1088/2632-2153/adaad8 |
| Open access | gold |
| Keywords | symbolic regression, neural networks, FPGA, pruning, high-energy physics, model compression |
Importance rating
real but narrower advance
Abstract states compact expressions are highly valuable for low-latency inference under stringent resource constraints at CERN LHC; important but a specific enabling technique.
More in Physics & Astronomy
| # | Paper | Year | Citations | Importance |
|---|---|---|---|---|
| 1 | Magnetic control of tokamak plasmas through deep reinforcement learning | 2022 | 739 |
7
|
| 2 | Solving the quantum many-body problem with artificial neural networks | 2017 | 2,205 |
7
|
| 3 | Learning high-accuracy error decoding for quantum processors | 2024 | 90 |
7
|
| 4 | Optimizing Quantum Error Correction Codes with Reinforcement Learning | 2019 | 158 |
7
|
| 5 | Machine-learning-assisted correction of correlated qubit errors in a topological code | 2018 | 141 |
7
|