Building a Machine Learning Accelerator with Silicon Dangling Bonds: From Verilog to Quantum Dot Layout
Bibliographic record
Abstract
At a time when traditional CMOS technologies approach their fundamental scaling limits and artificial intelligence continues to escalate global computational demands, emerging post-CMOS technologies like Silicon Dangling Bonds (SiDBs) provide promising pathways towards energy-efficient computation. SiDBs offer atomic-scale precision and discrete charge control, enabling the realization of ultra-dense computational logic. However, manual layout design and verification have historically restricted the exploration and scalability of SiDB-based logic systems. To this end, this work demonstrates an automated, end-to-end Electronic Design Automation (EDA) flow for designing and synthesizing a core component of a Matrix Multiply Unit (MXU) from high-level Register-transfer Level (RTL) Verilog descriptions down to dot-accurate SiDB layouts. Leveraging recent advances in SiDB-focused EDA tooling, we demonstrate the first fully automated design flow capable of translating RTL descriptions into manufacturable quantum-dot layouts. The proposed hierarchical Verilog approach addresses existing EDA constraints while facilitating comprehensive operational verification via test benches. Additionally, our design process incorporates reliability-focused Figures Of Merit (FoMs), ensuring the selection of robust logic gates throughout synthesis. Our synthesized MXU Processing Element (PE) layout represents a significant milestone in SiDB logic design, bridging previously manuallyintensive workflows with scalable, automated methodologies. Despite achieving larger footprints than hand-crafted designs, the presented approach provides a valuable foundation for future optimization and widespread adoption of SiDB-based computing architectures.
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 | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 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) | 0.006 | 0.002 |
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".