A Comparative Analysis of Off-Policy DRL Strategies for Analog Circuit Optimization: A Case Study on Bandgap References
Bibliographic record
Abstract
The optimization of analog circuits remains a challenging and time-intensive task, requiring expert knowledge to balance conflicting performance specifications. In this work, we explore the use of deep reinforcement learning (DRL) for automating analog circuit design, focusing on the comparative evaluation of the off-policy algorithms Soft Actor-Critic (SAC) and Deep Deterministic Policy Gradient (DDPG). We also analyze the impact of different reward functions and exploration strategies on convergence speed, consistency, and solution quality using a bandgap reference circuit as a test case. The primary objective is to minimize reference voltage variations across a temperature range of -40°C to 125°C to below 1 mV. Our findings show that SAC, particularly when paired with a structured reward function, achieves superior stability and robustness compared to DDPG, while the latter suffers from performance variability. The results highlight the crucial role of reward design in RL-based optimization and provide a guide for selecting effective strategies for analog circuit automation.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.005 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".