# Learning to Design Circuits

arXiv: https://arxiv.org/abs/1812.02734 · HTML: https://mlsanity.com/paper/1812.02734/

## Key facts

- Cited by 17 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 5 Dec 2018.

## Built on by (most cited first, 17 of 17)

- [QuantumNAS: Noise-Adaptive Search for Robust Quantum Circuits](https://mlsanity.com/paper/2107.10845/) (2021)
- [Machine Learning for Electronic Design Automation: A Survey](https://mlsanity.com/paper/2102.03357/) (2021)
- [Enable Deep Learning on Mobile Devices: Methods, Systems, and Applications](https://mlsanity.com/paper/2204.11786/) (2022)
- [CIRCUIT: A Benchmark for Circuit Interpretation and Reasoning Capabilities of LLMs](https://mlsanity.com/paper/2502.07980/) (2025)
- [AnaFlow: Agentic LLM-based Workflow for Reasoning-Driven Explainable and Sample-Efficient Analog Circuit Sizing](https://mlsanity.com/paper/2511.03697/) (2025)
- [Learning to Design Analog Circuits to Meet Threshold Specifications](https://mlsanity.com/paper/2307.13861/) (2023)
- [AICircuit: A Multi-Level Dataset and Benchmark for AI-Driven Analog Integrated Circuit Design](https://mlsanity.com/paper/2407.18272/) (2024)
- [Deep reinforcement learning-designed radiofrequency waveform in MRI](https://mlsanity.com/paper/2105.03061/) (2021)
- [ArchGym: An Open-Source Gymnasium for Machine Learning Assisted Architecture Design](https://mlsanity.com/paper/2306.08888/) (2023)
- [RobustAnalog: Fast Variation-Aware Analog Circuit Design Via Multi-task RL](https://mlsanity.com/paper/2207.06412/) (2022)
- [FALCON: An ML Framework for Fully Automated Layout-Constrained Analog Circuit Design](https://mlsanity.com/paper/2505.21923/) (2025)
- [AutoSizer: Automatic Sizing of Analog and Mixed-Signal Circuits via Large Language Model (LLM) Agents](https://mlsanity.com/paper/2602.02849/) (2026)
- [Boosting Pruned Networks with Linear Over-parameterization](https://mlsanity.com/paper/2204.11444/) (2022)
- [Automated Circuit Sizing with Multi-objective Optimization based on Differential Evolution and Bayesian Inference](https://mlsanity.com/paper/2206.02391/) (2022)
- [A Survey and Perspective on Artificial Intelligence for Security-Aware Electronic Design Automation](https://mlsanity.com/paper/2204.09579/) (2022)
- [pcbGPT: Automatic PCB Schematic Synthesis from Natural Language Requirements](https://mlsanity.com/paper/2606.01188/) (2026)
- [Fast Design Space Exploration of Nonlinear Systems: Part II](https://mlsanity.com/paper/2104.02464/) (2021)

## Abstract (arXiv)

Analog IC design relies on human experts to search for parameters that satisfy circuit specifications with their experience and intuitions, which is highly labor intensive, time consuming and suboptimal. Machine learning is a promising tool to automate this process. However, supervised learning is difficult for this task due to the low availability of training data: 1) Circuit simulation is slow, thus generating large-scale dataset is time-consuming; 2) Most circuit designs are propitiatory IPs within individual IC companies, making it expensive to collect large-scale datasets. We propose Learning to Design Circuits (L2DC) to leverage reinforcement learning that learns to efficiently generate new circuits data and to optimize circuits. We fix the schematic, and optimize the parameters of the transistors automatically by training an RL agent with no prior knowledge about optimizing circuits. After iteratively getting observations, generating a new set of transistor parameters, getting a reward, and adjusting the model, L2DC is able to optimize circuits. We evaluate L2DC on two transimpedance amplifiers. Trained for a day, our RL agent can achieve comparable or better performance than human experts trained for a quarter. It first learns to meet hard-constraints (eg. gain, bandwidth), and then learns to optimize good-to-have targets (eg. area, power). Compared with grid search-aided human design, L2DC can achieve 250boldsymbol× higher sample efficiency with comparable performance. Under the same runtime constraint, the performance of L2DC is also better than Bayesian Optimization.