Pith. sign in

REVIEW 4 major objections 4 minor 47 references

Mapping Fusion: Improving FPGA Technology Mapping with ASIC Mapper

T0 review · 4 major / 4 minor · reviewed 2026-08-06 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read FuseMap claims ASIC cell mapping run before FPGA LUT mapping cuts LUT count by 8% and area-delay product by 9%.

desk verdict The fused-mapping observation is new and worth taking seriously, but the RL-specific 8-9% claim is not supported by the experiments as reported. read the letter →

arxiv 2507.10912 v1 pith:F6SPE6DA submitted 2025-07-15 cs.AR

Cunxi Yu This is my paper · ORCID
classification cs.AR
keywords FPGAtechnologymappingLUTminimizationASICstandard-cellfusedflowmulti-armedbanditreinforcementlearningarea-delayproductlogicsynthesis
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

FPGA designers turn a Boolean network into LUTs — configurable blocks that each compute any function of K inputs, typically 4 to 6 — and the count and delay of those LUTs set the chip's area and speed. This paper claims that running an ASIC-style standard-cell mapper before LUT mapping, and feeding the resulting netlist into the FPGA mapper, consistently improves the FPGA result — but only if the ASIC cell library is a partial, design-specific subset, since the full library makes things worse. FuseMap finds that subset with a lightweight reinforcement learning loop: two multi-armed bandit agents repeatedly sample subsets of the library, run the fused mapping, and update cell-selection probabilities from the resulting LUT count, delay, or their product. Across ISCAS 85/89/99, VTR 8.0, and EPFL benchmark suites, the paper reports average gains of about 8% in LUT count and 9% in area-delay product, at the cost of at most 15 minutes of tuning per design. If correct, the result adds a new lever to FPGA logic synthesis that changes only the design recipe, not the FPGA architecture.

What carries the argument

The load-bearing mechanism is a fused two-stage mapping flow: ASIC-style technology mapping (the area-driven map -a command in the synthesis framework ABC) applied to the optimized And-Inverter Graph (a Boolean network of two-input AND gates with inverters on edges), followed by LUT mapping (if -K 6 with mfs2 post-optimization). Around this flow, FuseMap wraps a multi-armed bandit (MAB) library tuner, a lightweight reinforcement learning scheme with two agents: $\epsilon$-greedy, which chooses the currently best-scoring cell with probability $1-\epsilon$ and a random cell otherwise, and UCB, which adds an exploration bonus $c\sqrt{\log(t)/n_{a_i}}$ to each cell's score. Each agent keeps a probability vector $p$ over the $N$ cells of the standard-cell library, samples a subset of $n$ cells per iteration, and updates $p_{a_i}(t+1) = (p_{a_i}(t)n_{a_i}(t) + R_S(t))/n_a(t)$, where $R_S$ is the negative of the normalized reward — LUT count, delay, or their product, ADP, each scaled by the full-library baseline. The reward is pluggable, so the same tuner could target predicted post-routing timing or congestion. Reported per-design tuning runs finish within 15 minutes and typically converge in 10-12 iterations.

What would settle it

On the same benchmarks, run the fused mapping with the same 25-iteration budget and subset sizes but replace FuseMap's bandit with a sampler that draws cell subsets uniformly at random each round (or keeps only the best random subset found so far). If random sampling reproduces the average 8% LUT reduction and 9% ADP improvement, the reinforcement-learning component is not carrying the result. A weaker check: compare the full-library baseline against the average reward across all 25 iterations per design, not just the best iteration; an average that fails to beat the baseline would show the reported gain comes from cherry-picking the best round.

Watch

Extended reading notes

Core claim

The paper's central discovery is a fusion effect: ASIC standard-cell mapping can serve as a pre-packing phase that improves FPGA LUT mapping, but only when the standard-cell library is deliberately narrowed to a design-specific subset. The motivating experiments show three things: fusing in the full-library ASIC mapper raises LUT count by 7-24% compared with plain LUT mapping; randomly sampled partial libraries can beat both the vanilla fused flow and the plain mapper, by up to 30% on individual designs; yet the same random samples typically worsen average results, so the right subset must be found per design. FuseMap automates that search with two bandit agents ($\epsilon$-greedy and UCB) that keep a probability vector over the library's $N$ cells, sample a subset of $n$ cells, run the fused mapping, and update probabilities from a normalized reward based on LUT count, delay, or their product (ADP). The paper reports about 9% average LUT-count reduction, 3% delay reduction, and 9% ADP improvement across 24 designs with a 7nm library (the conclusions round the LUT gain to 8%), with the 9% ADP gain reproduced over a larger 161-design sweep; switching to a 45nm library changes the pre-LUT netlist structure and the results, showing the library choice itself matters.

Load-bearing premise

The paper reports the best result each design reached during up to 25 rounds of tuning; its case depends on that best number showing the learning algorithm's skill, not just the luck of trying many random cell subsets.

Editorial extensions

If this is right

  • A partial standard-cell library can outperform the full library as a pre-LUT mapping stage, so shrinking the ASIC library — normally expected to hurt mapping quality — can improve the downstream FPGA result for a specific design.
  • The gain is not tied to one technology: the same fused flow improves results with both 7nm and 45nm standard-cell libraries, suggesting the effect transfers across process design kits.
  • Per-design tuning is cheap — runs finish within 15 minutes and converge in about 10-12 iterations — so the search can be part of a normal synthesis run rather than a one-off offline optimization.
  • Because the reward is a configurable formula (LUT count, delay, or the product of LUT count and delay), the same bandit machinery extends to other objectives such as predicted post-routing timing or congestion.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The headline numbers are best-of-search: for each design the paper keeps the best outcome over 25 iterations and sample sizes {30, 40, 50, 60, 70}. A uniform-random library sampler given the same budget would test whether the bandit's learned selection — rather than search effort — is what produces the gains; that comparison is not in the paper.
  • If the fusion effect holds, LUT mapping could adopt standard-cell-like structures as an intermediate representation without an explicit ASIC mapper, for instance by synthesizing a synthetic cell library tuned to produce LUT-friendly netlists — a direction the paper notes but does not build.
  • A natural next experiment is transfer: whether the tuned library for one design helps similar designs, which would spread the per-design search cost across a benchmark suite or design family.
  • The delay metric here is LUT depth on the mapped network; carrying the fused flow through placement and routing would test whether the LUT-count and depth gains survive into wire delay and congestion.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 4 minor

Summary. The paper proposes FuseMap, a framework that inserts an ASIC standard-cell technology mapping step before FPGA LUT mapping and uses a multi-armed bandit (MAB) approach—specifically epsilon-greedy and UCB agents—to select a per-design subset of the ASIC library. The reported results claim average improvements of 8% in LUT count and 9% in area-delay product (ADP) over vanilla ABC LUT mapping, evaluated on benchmarks from ISCAS85/89, ITC/ISCAS99, VTR, and EPFL, with convergence plots and a comparison across two technology libraries. The central claim is that RL-guided cell selection makes fused ASIC-then-LUT mapping consistently better than vanilla LUT mapping.

Significance. If the claims are validated, the fused-mapping observation is genuinely interesting: it suggests that a partial standard-cell library can act as a pre-packing restructuring step that benefits LUT mapping, without changing the FPGA architecture. The paper makes source code available and uses standard benchmarks and libraries, which is a concrete strength. However, the current evidence does not isolate the RL contribution. The reported results are explicitly best-of-search values over 25 iterations and five sample sizes, and no equal-compute random-sampling baseline is provided. The motivating case study itself shows that random sampling can already produce up to 30% LUT reduction on some designs, so the claimed 8-9% averages may reflect search effort rather than learned cell-selection policy. The RL-specific contribution is therefore unproven, although the fused-mapping direction may still be worth pursuing.

major comments (4)
  1. [Section V (Results)] The reported FuseMap results are selected as the best over 25 iterations and sample sizes {30,40,50,60,70} with batch size 10, as stated in Section V: 'we evaluate the mapped results with different sampling sizes in FuseMap, which searches for the best achievable results.' No random-sampling baseline is reported with the same number of mapping calls or the same best-of-search protocol. Figure 1 already demonstrates that random sampling of 80/161 ASAP7 cells can reduce #LUT by up to 30% on s832a, while average results are worse. Thus the average 8-9% improvements over vanilla LUT mapping could be an artifact of the number of random trials rather than evidence that the MAB policy learns design-specific cell selections. Please add an equal-budget random subset baseline and report mean, median, and dispersion over independent runs.
  2. [Section IV (Approach, Reward update)] The reward function is defined as RS = -(DS/DBase) * (AS/ABase), where DBase and ABase are 'established using all the cells in the original library for technology mapping' (i.e., full-library ASIC mapping). Section V, however, reports all improvements relative to vanilla LUT mapping. This is a mismatch between the objective being optimized and the metric being reported. A subset that maximizes reward relative to full-library ASIC mapping is not necessarily better than vanilla LUT mapping, so the reported gains may not correspond to the intended optimization target. Please either redefine the reward baseline as vanilla LUT mapping or report results against the same baseline used in the reward.
  3. [Section V (RQ1)] The claim of 'consistent improvements' is supported only by Figure 3, which shows normalized area, delay, and ADP per design with no error bars, no mean or median over independent runs, and no per-design table. The abstract and Section VI state an average 8% LUT improvement, while Section V states approximately 9%; these numbers must be reconciled. In addition, the statement that the full ISCAS89/EPFL/VTR suites (161 designs) confirm 9% ADP improvement is not backed by any figure, table, or detailed data in the manuscript.
  4. [Section IV-B, Eq. (2)] The probability update equation is malformed as printed: pai(t+1) = pai(t)nai(t) + RS(t)na(t) is dimensionally inconsistent, and the denominator is missing or ambiguous. Given that this equation is the core learning update for both FuseMap-epsilon and FuseMap-UCB, it must be corrected and the notation for n_a(t) defined precisely. Without a correct update, the algorithm cannot be reproduced from the description.
minor comments (4)
  1. [Section V (Results)] There are typos in Section V: 'libary' should be 'library' and 'invovles' should be 'involves'.
  2. [Abstract and Section V] The abstract and conclusions give 8% average LUT improvement, while Section V RQ1 states approximately 9%; please make the reported numbers consistent.
  3. [Figure 3] Figure 3 does not specify whether each point is the best-of-search result, the final iteration result, or an average over runs, nor does it distinguish FuseMap-epsilon from FuseMap-UCB. Please state this clearly in the caption or text.
  4. [Table I and Section IV] The notation p_ai and a_i is typeset inconsistently; please use a consistent subscript style and define all symbols in Table I.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the reported gains are measured empirical outcomes of a bandit search, not quantities defined from the claimed result; the MapTune self-citation is not load-bearing.

full rationale

FuseMap's chain from formulation to result is an empirical search, not a derivation in which the output is equal to an input by construction. The MAB update and reward equations in Section IV define how cell-selection probabilities are updated, but the reported LUT/delay/ADP numbers in Section V are direct mapping results from ABC, not values reconstructed from the reward function. The reward normalization uses DBase and ABase from full-library ASIC mapping while the headline comparison uses vanilla LUT mapping; this is an objective-alignment weakness, but the reported 8-9% improvements are not forced by that normalization. Section V also states that FuseMap 'searches for the best achievable results' over sample sizes {30,40,50,60,70}; reporting best-of-search numbers without an equal-budget random baseline weakens the attribution of gains to the learned policy, but that is a benchmarking limitation, not a circular reduction. The reuse of MapTune [29] is a true self-citation, yet it is not load-bearing: the paper supplies its own MAB formulation and measures results against external ABC baselines on ISCAS/VTR/EPFL benchmarks, and no uniqueness theorem or ansatz is imported from prior work to forbid alternatives. Accordingly, no circular step can be quoted, and the paper is self-contained with respect to the circularity question.

Assumptions & free parameters 4 free parameters · 4 assumptions · 0 invented entities

The method introduces no new physical or algorithmic entities beyond subsets of existing standard-cell libraries. Its empirical claims rest on hand-set hyperparameters, on the correctness of ABC's internal transformations, and on the assumption that best-of-N search results represent the learned policy. No independent falsifiable prediction is made beyond the benchmark QoR numbers.

free parameters (4)
  • epsilon-greedy exploration probability (epsilon)
    Controls exploration vs exploitation in FuseMap-epsilon; no value or tuning method is reported in Section IV-B.
  • UCB exploration coefficient c
    Scales the confidence bound in Eq 1; exact value is not reported, so results are not reproducible without contacting the authors.
  • sample size n (subset size) = {30, 40, 50, 60, 70}
    The number of cells selected from the library per iteration is fixed by hand; results and convergence depend strongly on this choice (Section V RQ2).
  • iteration limit and batch size = 25 iterations, batch size 10
    Termination and sampling batch are user-defined constants in Section V; no sensitivity analysis is given.
assumptions (4)
  • domain assumption ABC's map, if -K, and mfs2 preserve functional equivalence and can be chained after an ASIC map step followed by strash/resyn.
    All experiments run inside ABC; the paper never verifies equivalence of the fused mapped netlists (Section IV-V).
  • domain assumption ASAP7 and FreePDK45 library area/delay models are faithful enough that LUT count and depth after ABC mapping predict FPGA QoR.
    Reported area and delay metrics come from these standard-cell libraries rather than from FPGA place-and-route (Sections IV-V).
  • ad hoc to paper The best result over multiple MAB runs and sample sizes is a valid performance estimate of the FuseMap approach.
    Section V states FuseMap 'searches for the best achievable results'; this protocol is not justified against a random-search baseline.
  • standard math The probability update rule in Eq 2 correctly estimates each cell's contribution to reward.
    The paper uses the average reward update without proof of convergence; it is a heuristic bandit update common in the literature.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Mapping Fusion: Improving FPGA Technology Mapping with ASIC Mapper." pith.science (2026). https://pith.science/paper/F6SPE6DA

@misc{pith2026250710912,
  author       = {Pith},
  title        = {Pith review of: Mapping Fusion: Improving FPGA Technology Mapping with ASIC Mapper},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/F6SPE6DA}},
  note         = {Machine review of arXiv:2507.10912}
}
read the original abstract

LUT (Look-Up Table) mapping is a critical step in FPGA logic synthesis, where a logic network is transformed into a form that can be directly implemented using the FPGA's LUTs. An FPGA LUT is a flexible digital memory structure that can implement any logic function of a limited number of inputs, typically 4 to 6 inputs, depending on the FPGA architecture. The goal of LUT mapping is to map the Boolean network into LUTs, where each LUT can implement any function with a fixed number of inputs. In parallel to FPGA technology mapping, ASIC technology mapping maps the Boolean network to user-defined standard cells, which has traditionally been developed separately from LUT mapping algorithms. However, in this work, our motivating examples demonstrate that ASIC technology mappers can potentially improve the performance of LUT mappers, such that standard cell mapping and LUT mapping work in an incremental manner. Therefore, we propose the FuseMap framework, which explores this opportunity to improve LUT mapping in the FPGA design flow by utilizing reinforcement learning to make design-specific choices during cell selection. The effectiveness of FuseMap is evaluated on a wide range of benchmarks, different technology libraries, and technology mappers. The experimental results demonstrate that FuseMap achieves higher mapping accuracy while reducing delay and area across diverse circuit designs collected from ISCAS 85/89, ITC/ISCAS 99, VTR 8.0, and EPFL benchmarks.

Figures

Figures reproduced from arXiv: 2507.10912 by the authors.

Figure 1
Figure 1. Motivation case studies of fusing ASIC and LUT mapping [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Overview of the FuseMap Framework. Termination is based on [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. FuseMap results with three different reward functions over 24 designs using ASAP7 library. [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Convergence analysis of FuseMap with #LUT size as targeted minimization objective. [PITH_FULL_IMAGE:figures/full_fig_p005_4.png]
Figure 5
Figure 5. Figure 5: FuseMap performance comparisons between ASAP7 (baseline=1) and NAN45 libraries. [PITH_FULL_IMAGE:figures/full_fig_p006_5.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

47 extracted references · 45 canonical work pages

  1. [1]

    On area/depth trade-off in lut-based fpga technology mapping

    Jason Cong and Yuzheng Ding. On area/depth trade-off in lut-based fpga technology mapping. In Proceedings of the 30th International Design Automation Conference, pages 213–218, 1993

    work page 1993

  2. [2]

    Flowmap: An optimal technology mapping algorithm for delay optimization in lookup-table based fpga designs

    Jason Cong and Yuzheng Ding. Flowmap: An optimal technology mapping algorithm for delay optimization in lookup-table based fpga designs. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems , 13(1):1–12, 1994

    work page 1994

  3. [3]

    Complexity of the lookup- table minimization problem for fpga technology mapping

    Amir H Farrahi and Majid Sarrafzadeh. Complexity of the lookup- table minimization problem for fpga technology mapping. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 13(11):1319–1332, 1994

    work page 1994

  4. [4]

    Low-power technology mapping for fpga architectures with dual supply voltages

    Deming Chen, Jason Cong, Fei Li, and Lei He. Low-power technology mapping for fpga architectures with dual supply voltages. In Proceedings of the 2004 ACM/SIGDA 12th international symposium on Field programmable gate arrays, pages 109–117, 2004

    work page 2004

  5. [5]

    Improvements to technology mapping for lut-based fpgas

    Alan Mishchenko, Satrajit Chatterjee, and Robert Brayton. Improvements to technology mapping for lut-based fpgas. In Proceedings of the 2006 ACM/SIGDA 14th international symposium on Field programmable gate arrays, pages 41–49, 2006

    work page 2006

  6. [6]

    Fpga technology mapping: a study of optimality

    Andrew Ling, Deshanand P Singh, and Stephen D Brown. Fpga technology mapping: a study of optimality. In Proceedings of the 42nd annual Design Automation Conference , pages 427–432, 2005

    work page 2005

  7. [7]

    Technology mapping for fpgas with embedded memory blocks

    Jason Cong and Songjie Xu. Technology mapping for fpgas with embedded memory blocks. In Proceedings of the 1998 ACM/SIGDA sixth international symposium on Field programmable gate arrays , pages 179–188, 1998

    work page 1998

  8. [8]

    Combinational profiles of sequential benchmark circuits

    Franc Brglez, David Bryan, and Krzysztof Kozminski. Combinational profiles of sequential benchmark circuits. In 1989 IEEE International Symposium on Circuits and Systems (ISCAS) , pages 1929–1934. IEEE, 1989

    work page 1989

Show all 47 references
  1. [9]

    VTR 8.0: Next generation architecture and CAD system for FPGAs

    Jason Luu, Jeffrey Goeders, Michael Wainberg, Andrew Somerville, Thien Yu, Konstantin Nasartschuk, Miad Nasr, Sen Wang, Tim Liu, Nooruddin Ahmed, et al. VTR 8.0: Next generation architecture and CAD system for FPGAs. ACM Transactions on Reconfigurable Technology and Systems (T...

    2014

  2. [10]

    The epfl logic synthesis libraries

    Mathias Soeken, Heinz Riener, Winston Haaswijk, Eleonora Testa, Bruno Schmitt, Giulia Meuli, Fereshte Mozafari, and Giovanni De Micheli. The epfl logic synthesis libraries. arXiv preprint arXiv:1805.05121 , 2018

    2018 arXiv

  3. [11]

    Tree-based mapping of algorithms to predefined structures

    Peter Marwedel. Tree-based mapping of algorithms to predefined structures. In Proceedings of 1993 International Conference on Computer Aided Design (ICCAD) , pages 586–593. IEEE, 1993

    1993

  4. [12]

    Technology Mapping with Boolean Matching, Supergates and Choices

    Alan Mishchenko, Satrajit Chatterjee, Robert Brayton, Xinning Wang, and Timothy Kam. Technology Mapping with Boolean Matching, Supergates and Choices. 2005

    2005

  5. [13]

    Alan Mishchenko, Satrajit Chatterjee, and Robert K. Brayton. Dag- aware AIG rewriting a fresh look at combinational logic synthesis. In Proceedings of the 43rd Design Automation Conference, DAC 2006, San Francisco, CA, USA, July 24-28, 2006 , pages 532–535, 2006

    2006

  6. [14]

    Read your circuit: leveraging word embedding to guide logic optimization

    Walter Lau Neto, Matheus Trevisan Moreira, Luca Amaru, Cunxi Yu, and Pierre-Emmanuel Gaillardon. Read your circuit: leveraging word embedding to guide logic optimization. In Proceedings of the 26th Asia and South Pacific Design Automation Conference , pages 530–535, 2021

    2021

  7. [15]

    Boolgebra: Attributed graph-learning for boolean algebraic manipulation

    Yingjie Li, Anthony Agnesina, Yanqing Zhang, Haoxing Ren, and Cunxi Yu. Boolgebra: Attributed graph-learning for boolean algebraic manipulation. In 2024 Design, Automation & Test in Europe Conference & Exhibition (DATE) , pages 1–2. IEEE, 2024

    2024

  8. [16]

    Slap: A supervised learning approach for priority cuts technology mapping

    Walter Lau Neto, Matheus T Moreira, Yingjie Li, Luca Amar `u, Cunxi Yu, and Pierre-Emmanuel Gaillardon. Slap: A supervised learning approach for priority cuts technology mapping. In 2021 58th ACM/IEEE Design Automation Conference (DAC), pages 859–864. IEEE, 2021

    2021

  9. [17]

    Nvcell: Standard cell layout in advanced technology nodes with reinforcement learning

    Haoxing Ren and Matthew Fojtik. Nvcell: Standard cell layout in advanced technology nodes with reinforcement learning. In 2021 58th ACM/IEEE Design Automation Conference (DAC) , pages 1291–1294. IEEE, 2021

    2021

  10. [18]

    InTime: A Machine Learning Approach for Efficient Selection of FPGA CAD Tool Parameters

    Nachiket Kapre, Harnhua Ng, Kirvy Teo, and Jaco Naude. InTime: A Machine Learning Approach for Efficient Selection of FPGA CAD Tool Parameters. February 2015

    2015

  11. [19]

    LAMDA: Learning-Assisted Multi-stage Autotuning for FPGA Design Closure

    Ecenur Ustun, Shaojie Xiang, Jinny Gui, Cunxi Yu, and Zhiru Zhang. LAMDA: Learning-Assisted Multi-stage Autotuning for FPGA Design Closure. In FCCM’19

  12. [20]

    Ziegler, Ramon Bertran Monfort, Alper Buyuktosunoglu, and Pradip Bose

    Matthew M. Ziegler, Ramon Bertran Monfort, Alper Buyuktosunoglu, and Pradip Bose. Machine Learning Techniques for Taming the Complexity of Modern Hardware Design. IBM Journal of Research and Development, 61(4):13, 2017

    2017

  13. [21]

    Developing synthesis flows without human knowledge

    Cunxi Yu, Houping Xiao, and Giovanni De Micheli. Developing synthesis flows without human knowledge. In Proceedings of the 55th Annual Design Automation Conference, DAC 2018, San Francisco, CA, USA, June 24-29, 2018 , pages 50:1–50:6, 2018

    2018

  14. [22]

    Drills: Deep reinforcement learning for logic synthesis

    Abdelrahman Hosny, Soheil Hashemi, Mohamed Shalan, and Sherief Reda. Drills: Deep reinforcement learning for logic synthesis. arXiv preprint arXiv:1911.04021, 2019

    1911 arXiv

  15. [23]

    High performance graph convolutional networks with applications in testability analysis

    Yuzhe Ma, Haoxing Ren, Brucek Khailany, Harbinder Sikka, Lijuan Luo, Karthikeyan Natarajan, and Bei Yu. High performance graph convolutional networks with applications in testability analysis. In Proceedings of the 56th Annual Design Automation Conference 2019 , pages 1–6, 2019

    2019

  16. [24]

    Decision making in synthesis cross technologies using lstms and transfer learning

    Cunxi Yu and Wang Zhou. Decision making in synthesis cross technologies using lstms and transfer learning. In Proceedings of the 2020 ACM/IEEE Workshop on Machine Learning for CAD , pages 55–60, 2020

    2020

  17. [25]

    Flowtune: Practical multi-armed bandits in boolean optimiza- tion

    Cunxi Yu. Flowtune: Practical multi-armed bandits in boolean optimiza- tion. In Proceedings of the 39th International Conference on Computer- Aided Design, pages 1–9, 2020

    2020

  18. [26]

    Flowtune: End-to-end automatic logic optimization exploration via domain-specific multi-armed bandit

    Walter Lau Neto, Yingjie Li, Pierre-Emmanuel Gaillardon, and Cunxi Yu. Flowtune: End-to-end automatic logic optimization exploration via domain-specific multi-armed bandit. IEEE Transactions on Computer- Aided Design of Integrated Circuits and Systems , 2022

    2022

  19. [27]

    Machine learning for agile fpga design

    Debjit Pal, Chenhui Deng, Ecenur Ustun, Cunxi Yu, and Zhiru Zhang. Machine learning for agile fpga design. Machine Learning Applications in Electronic Design Automation , pages 471–504, 2022

    2022

  20. [28]

    Efficient Design Space Exploration via Statistical Sampling and AdaBoost Learning

    Dandan Li, Shuzhen Yao, Yu-Hang Liu, Senzhang Wang, and Xian-He Sun. Efficient Design Space Exploration via Statistical Sampling and AdaBoost Learning. June 2016

    2016

  21. [29]

    Maptune: Advancing asic technology mapping via reinforcement learning guided library tuning

    Mingju Liu, Daniel Robinson, Yingjie Li, and Cunxi Yu. Maptune: Advancing asic technology mapping via reinforcement learning guided library tuning. In Proceedings of the 43rd IEEE/ACM International Conference on Computer-Aided Design , pages 1–10, 2024

    2024

  22. [30]

    Verilog-to- pyg-a framework for graph learning and augmentation on rtl designs

    Yingjie Li, Mingju Liu, Alan Mishchenko, and Cunxi Yu. Verilog-to- pyg-a framework for graph learning and augmentation on rtl designs. In 2023 IEEE/ACM International Conference on Computer Aided Design (ICCAD), pages 1–4. IEEE, 2023

    2023

  23. [31]

    E-syn: E-graph rewriting with technology-aware cost functions for logic synthesis

    Chen Chen, Guangyu Hu, Dongsheng Zuo, Cunxi Yu, Yuzhe Ma, and Hongce Zhang. E-syn: E-graph rewriting with technology-aware cost functions for logic synthesis. In Proceedings of the 61st ACM/IEEE Design Automation Conference , pages 1–6, 2024

    2024

  24. [32]

    Gamora: Graph learning based symbolic reasoning for large-scale boolean networks

    Nan Wu, Yingjie Li, Cong Hao, Steve Dai, Cunxi Yu, and Yuan Xie. Gamora: Graph learning based symbolic reasoning for large-scale boolean networks. In 2023 60th ACM/IEEE Design Automation Conference (DAC), pages 1–6. IEEE, 2023

    2023

  25. [33]

    Less is more: Hop-wise graph attention for scalable and generalizable learning on circuits

    Chenhui Deng, Zichao Yue, Cunxi Yu, Gokce Sarar, Ryan Carey, Rajeev Jain, and Zhiru Zhang. Less is more: Hop-wise graph attention for scalable and generalizable learning on circuits. In Proceedings of the 61st ACM/IEEE Design Automation Conference , pages 1–6, 2024

    2024

  26. [34]

    Smoothe: Differentiable e-graph extraction

    Yaohui Cai, Kaixin Yang, Chenhui Deng, Cunxi Yu, and Zhiru Zhang. Smoothe: Differentiable e-graph extraction. In Proceedings of the 30th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Volume 1 , pages 1020–1034, 2025

    2025

  27. [35]

    Differ- entiable combinatorial scheduling at scale

    Mingju Liu, Yingjie Li, Jiaqi Yin, Zhiru Zhang, and Cunxi Yu. Differ- entiable combinatorial scheduling at scale. In Proceedings of the 41st International Conference on Machine Learning , ICML’24. JMLR.org, 2024

    2024

  28. [36]

    Respect: Reinforcement learning based edge scheduling on pipelined coral edge tpus

    Jiaqi Yin, Yingjie Li, Daniel Robinson, and Cunxi Yu. Respect: Reinforcement learning based edge scheduling on pipelined coral edge tpus. In 2023 60th ACM/IEEE Design Automation Conference (DAC) , pages 1–6. IEEE, 2023

    2023

  29. [37]

    Accelerating exact combinatorial optimization via rl-based initialization-a case study in scheduling

    Jiaqi Yin and Cunxi Yu. Accelerating exact combinatorial optimization via rl-based initialization-a case study in scheduling. In 2023 IEEE/ACM International Conference on Computer Aided Design (ICCAD) , pages 1–9. IEEE, 2023

    2023

  30. [38]

    Cbtune: Contextual bandit tuning for logic synthesis

    Fangzhou Liu, Zehua Pei, Ziyang Yu, Haisheng Zheng, Zhuolun He, Tinghuan Chen, and Bei Yu. Cbtune: Contextual bandit tuning for logic synthesis. 2024

    2024

  31. [39]

    Abc: A system for sequential synthesis and verification

    Alan Mishchenko et al. Abc: A system for sequential synthesis and verification. http://www. eecs. berkeley. edu/alanmi/abc , 17, 2007

    2007

  32. [40]

    ABC: An Academic Industrial- strength Verification tool

    Robert Brayton and Alan Mishchenko. ABC: An Academic Industrial- strength Verification tool. In Computer Aided Verification, pages 24–40. Springer, 2010

    2010

  33. [41]

    Painting on placement: Forecasting routing congestion using conditional generative adversarial nets

    Cunxi Yu and Zhiru Zhang. Painting on placement: Forecasting routing congestion using conditional generative adversarial nets. In Proceedings of the 56th Annual Design Automation Conference 2019 , pages 1–6, 2019

    2019

  34. [42]

    Fast and accurate estimation of quality of results in high-level synthesis with machine learning

    Steve Dai, Yuan Zhou, Hang Zhang, Ecenur Ustun, Evangeline FY Young, and Zhiru Zhang. Fast and accurate estimation of quality of results in high-level synthesis with machine learning. In 2018 IEEE 26th Annual International Symposium on Field-Programmable Custom Computing Machi...

    2018

  35. [43]

    Robust gnn-based representation learning for hls

    Atefeh Sohrabizadeh, Yunsheng Bai, Yizhou Sun, and Jason Cong. Robust gnn-based representation learning for hls. In 2023 IEEE/ACM International Conference on Computer Aided Design (ICCAD) , pages 1–9. IEEE, 2023

    2023

  36. [44]

    Algorithms for multi-armed bandit problems

    V olodymyr Kuleshov and Doina Precup. Algorithms for multi-armed bandit problems. arXiv preprint arXiv:1402.6028 , 2014

    2014 arXiv

  37. [45]

    Upper confidence bound based decision making strategies and dynamic spectrum access

    Wassim Jouini, Damien Ernst, Christophe Moy, and Jacques Palicot. Upper confidence bound based decision making strategies and dynamic spectrum access. In 2010 IEEE International Conference on Communi- cations, pages 1–5. IEEE, 2010

    2010

  38. [46]

    Asap7: A 7-nm finfet predictive process design kit

    Lawrence T Clark, Vinay Vashishtha, Lucian Shifren, Aditya Gujja, Saurabh Sinha, Brian Cline, Chandarasekaran Ramamurthy, and Greg Yeric. Asap7: A 7-nm finfet predictive process design kit. Microelec- tronics Journal, 53:105–115, 2016

    2016

  39. [47]

    URL https://eda.ncsu.edu/freepdk/freepdk45/

    NC State FreePDK45’s Documentation. URL https://eda.ncsu.edu/freepdk/freepdk45/

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.