SPICEMixer - Netlist-Level Circuit Evolution
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915626382000128 |
|---|---|
| author | Uhlich, Stefan Bonetti, Andrea Venkitaraman, Arun Hsieh, Chia-Yu Gençer, Yağız Gürsoy, Mustafa Emre Matsuo, Ryoga Servadei, Lorenzo |
| author_facet | Uhlich, Stefan Bonetti, Andrea Venkitaraman, Arun Hsieh, Chia-Yu Gençer, Yağız Gürsoy, Mustafa Emre Matsuo, Ryoga Servadei, Lorenzo |
| contents | We present SPICEMixer, a genetic algorithm that synthesizes circuits by directly evolving SPICE netlists. SPICEMixer operates on individual netlist lines, making it compatible with arbitrary components and subcircuits and enabling general-purpose genetic operators: crossover, mutation, and pruning, all applied directly at the netlist level. To support these operators, we normalize each netlist by enforcing consistent net naming (inputs, outputs, supplies, and internal nets) and by sorting components and nets into a fixed order, so that similar circuit structures appear at similar line positions. This normalized netlist format improves the effectiveness of crossover, mutation, and pruning. We demonstrate SPICEMixer by synthesizing standard cells (e.g., NAND2 and latch) and by designing OpAmps that meet specified targets. Across tasks, SPICEMixer matches or exceeds recent synthesis methods while requiring substantially fewer simulations. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_01497 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | SPICEMixer - Netlist-Level Circuit Evolution Uhlich, Stefan Bonetti, Andrea Venkitaraman, Arun Hsieh, Chia-Yu Gençer, Yağız Gürsoy, Mustafa Emre Matsuo, Ryoga Servadei, Lorenzo Neural and Evolutionary Computing Hardware Architecture Machine Learning B.7.0 We present SPICEMixer, a genetic algorithm that synthesizes circuits by directly evolving SPICE netlists. SPICEMixer operates on individual netlist lines, making it compatible with arbitrary components and subcircuits and enabling general-purpose genetic operators: crossover, mutation, and pruning, all applied directly at the netlist level. To support these operators, we normalize each netlist by enforcing consistent net naming (inputs, outputs, supplies, and internal nets) and by sorting components and nets into a fixed order, so that similar circuit structures appear at similar line positions. This normalized netlist format improves the effectiveness of crossover, mutation, and pruning. We demonstrate SPICEMixer by synthesizing standard cells (e.g., NAND2 and latch) and by designing OpAmps that meet specified targets. Across tasks, SPICEMixer matches or exceeds recent synthesis methods while requiring substantially fewer simulations. |
| title | SPICEMixer - Netlist-Level Circuit Evolution |
| topic | Neural and Evolutionary Computing Hardware Architecture Machine Learning B.7.0 |
| url | https://arxiv.org/abs/2506.01497 |