Pith. sign in

REVIEW 3 major objections 6 minor 29 references

DiffPower: GPU-Accelerated Differentiable Switching Power Analysis and Optimization

T0 review · 3 major / 6 minor · reviewed 2026-08-05 · deepseek-v4-flash

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

Pith's one-line read DiffPower claims switching-power analysis can be made differentiable and GPU-fast—1,002x over CPU propagation—by compiling netlists to bytecode, calibrating with bit-packed simulation, and using the resulting power gradients to guide cell s

desk verdict DiffPower is a real engineering advance with a genuinely solid power-virus result, but the headline sizing gains are measured on its own approximate power model—worth a referee, not yet worth trusting for signoff claims. read the letter →

arxiv 2608.03778 v1 pith:CR4Y45QM submitted 2026-08-04 cs.AR cs.DC

Isaac Jacobson , Zheng Zhao , Rashmi Mehrotra , Guanglei Zhou , Vineet Rashingkar , Yiran Chen This is my paper · ORCID
classification cs.ARcs.DC
keywords switchingpoweranalysisGPUaccelerationautomaticdifferentiationstaticprobabilitypropagationpower-awarecellsizingvirusgenerationstraight-throughestimatorgate-levelsimulation
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

DiffPower claims that switching-power analysis can be made both fast and differentiable by compiling a netlist into a PDK-agnostic bytecode program, running static-probability propagation on a GPU, and calibrating the analytical model with a bit-packed simulation pass. This yields per-pin power gradients—sensitivities of total switching power to each pin's activity—at speeds that make them usable inside optimization loops, not just as a one-time report. On a 652K-cell design, forward and backward passes each finish in under 30 ms, roughly 1,000x faster than single-threaded CPU propagation, with toggle-rate correlations of r=0.96 against a commercial reference across ten designs. The paper demonstrates that these gradients improve two downstream tasks: gradient-weighted cell sizing beats local-power heuristics by up to 2.98x, and gradient-ascent power-virus search beats evolutionary search by up to 2.13x while cutting runtime from hours to under a minute on designs up to 117K cells. If true, the trade-off between analysis fidelity and optimization speed that has constrained power-aware physical design would be substantially relaxed.

What carries the argument

The load-bearing object is the per-pin power gradient, defined as the derivative of total switching power with respect to a pin's switching activity, obtained by running reverse-mode automatic differentiation backwards through a levelized pin-level DAG. The DAG is evaluated as a stack-machine bytecode program: each cell function compiles to ops (IN, NOT, AND, OR, XOR, CONST) that the same interpreter executes forward for static-probability propagation and backward for exact derivatives; common gates get hand-optimized kernels. To close the fidelity gap, a bit-packed 64-bit parallel simulation pass computes an accurate toggle rate per pin, and the ratio TR_acc/TR_model, clipped to [0.1,10], s

What would settle it

Run the same sizing and power-virus flows on a new design while replacing DiffPower's gradients with finite-difference gradients of the commercial signoff power (or a high-fidelity vector simulation), and compare optimization outcomes. If beta-scaled gradients do not rank pins the same way as signoff gradients on a design with heavy reconvergence—the failure mode already visible at r=0.84 on the 652K design—the 2.98x sizing and 2.13x power-virus gains would not reproduce, and the straight-through estimator would be the cause to isolate by ablating beta.

Watch

Extended reading notes

Core claim

The paper's central claim is that a differentiable surrogate for switching power—built by treating every gate as a Boolean probability function and every wire as a probability copy—can be made accurate enough and fast enough to guide real optimization. The mechanism is a levelized pin-level DAG resident in GPU memory, with Boolean cell functions compiled to stack-machine bytecode; the same bytecode runs forward static-probability propagation and reverse-mode automatic differentiation, so arbitrary cell libraries get gradients with no gate-specific derivative code. A second, bit-packed 64-bit simulation pass measures per-pin toggle rates and computes scaling factors beta_i that multiply the a

Load-bearing premise

The load-bearing premise is that the beta-scaled analytical gradients—beta estimated from a 100-cycle bit-packed simulation and held constant during backprop—point in directions that improve true switching power on real designs; the paper validates gradients against finite differences of its own surrogate, not against signoff power gradients, so a design where this estimator misleads the optimizer would break both downstream claims.

Editorial extensions

If this is right

  • Switching-power analysis can move inside iterative optimization loops: re-running full forward/backward passes on a 652K-cell design takes tens of milliseconds, so gradients can be refreshed every few resize decisions instead of computed once.
  • New cell libraries need no new derivative code: any Boolean function compiles to bytecode, and both propagation and backprop follow automatically, easing retargeting across technology nodes.
  • Cell sizing can be driven by global sensitivity rather than local power, so downsizing choices that cut downstream load are found even where local heuristics plateau, as on the 117K-cell design.
  • Power-virus search becomes a fixed-cost gradient ascent with T=100 steps, replacing population-based search; on the evaluated designs it dominates evolutionary search in both score and runtime.
  • Gradient rank agreement with numerical differentiation (top-500 Jaccard 1.0 on the medium design) means the model's ordering of high-impact pins is trustworthy enough to prioritize optimization effort.

Reading between the lines

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

  • The beta-calibration scheme is presented as a fixed, parameter-free scaling; an implicit extension is to fit beta or a small correction function from a few simulation vectors, potentially recovering toggle-rate fidelity on the large 652K design where r=0.84.
  • Because gradients capture downstream fanout effects, the same machinery could be applied to other power-relevant discrete choices, such as logic restructuring, pin swapping, or multi-voltage assignment, not just cell sizing.
  • Power-virus search could be hybridized: use gradient ascent to initialize a small set of candidates, then apply one-bit flip refinement; the paper notes the smallest designs show the narrowest gap, and a discrete local polish would likely close it.
  • The combinational-boundary scoping is the main accuracy limit; extending the levelized graph to treat sequential feedback as differentiable state, or modeling clock and infrastructure nets explicitly, is the natural next regime the architecture already points toward.
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

3 major / 6 minor

Summary. DiffPower presents a GPU-resident, levelized pin-graph engine that computes static-probability (SP) propagation and toggle-rate (TR) estimates through a hybrid of algebraic SP analysis and bit-packed simulation. A bytecode representation of arbitrary Boolean cells enables reverse-mode automatic differentiation with no gate-specific derivative code. The paper reports up to 1,002× speedup over single-threaded CPU propagation, a median TR correlation of r=0.96 against a commercial reference, near-perfect rank agreement between analytical and finite-difference gradients of its own surrogate, and two downstream applications: gradient-weighted cell sizing (claimed 2.98× improvement over local-power heuristics) and gradient-ascent power virus generation (up to 2.13× higher transition-weighted power than evolutionary search).

Significance. If the claims hold, DiffPower is a significant contribution to EDA: it extends differentiable gate-level analysis to switching power at industrial scale, provides a PDK-agnostic bytecode abstraction, and demonstrates a fast power-virus generator that is evaluated on the true discrete transition-weighted metric (Eq. 3) rather than on a proxy. The accuracy comparison against an external commercial tool is a genuine strength, as is the use of the same P_virus metric for the evolutionary baseline. The main gap is that the headline cell-sizing result is presented as power reduction but appears to be computed by DiffPower's own approximate model; without a signoff-power evaluation, the 2.98×/4.4× ratios are not evidence about real design power. The framework's speed and architectural ideas are likely to be of interest regardless, but the practical optimization claims need strengthening.

major comments (3)
  1. [§IV-A, Fig. 3] The claimed 2.98×/4.4× sizing improvements are not substantiated by signoff-level power measurements. In §IV-A the acceptance criterion is 'reduces switching power', but the only commercial query explicitly described is slack from the commercial timer; no statement says the power reduction plotted in Fig. 3 is computed by a commercial power engine. If the y-axis is DiffPower's own Eq. (1)–(2) (optionally β-scaled via Eq. (5)–(6)), then the experiment optimizes and evaluates the same approximate objective, so a gradient-guided method is expected to beat local heuristics on that model. Please add a final signoff-power evaluation of the sized netlists, or at least report separately the DiffPower-model reduction and the signoff reduction. The current text does not support the external validity of the core sizing claim.
  2. [§V-C] The finite-difference validation establishes that the backward pass correctly differentiates the algebraic SP forward model, not that the resulting gradients correlate with signoff switching-power sensitivity. The large-scale rank comparison (Jaccard=1.0, Pearson≈1.0) is between analytical gradients and CPU finite differences of the same surrogate. Since the downstream sizing and virus generation rely on these gradients, an important missing link is a comparison of top gradient-ranked pins/cells against a commercial power analysis engine (or at least against the hybrid TR correlation of Table III). As written, the 'near-perfect' gradient claim supports implementation correctness only; the optimization-quality claims must be carried by the downstream experiments, which are themselves incomplete for sizing (see previous comment).
  3. [§III-D, Eq. (6), §V-C] Eq. (6) scales backward gradients by β_i treated as a constant, whereas β_i in Eq. (5) is computed from the simulated toggle rate and therefore depends on the input probabilities. The resulting backward pass is a straight-through estimator, not the exact gradient of the hybrid forward model. The validation in §V-C is described as testing 'derivatives of the forward SP model'—apparently the algebraic model without β-scaling. If hybrid (β-scaled) gradients are used in the sizing metric or in Algorithm 4, please state this explicitly and validate the hybrid backward pass under fixed β, or clearly delimit which experiments use unscaled algebraic gradients. Without this clarification, it is unclear whether the finite-difference validation covers the deployed gradient.
minor comments (6)
  1. [Table II] The table's last column is labeled 'GPU vs. CPU', but the text also claims 'three to four orders of magnitude faster than the commercial reference tool'. Add an explicit GPU vs. Ref. column or report those ratios in the text.
  2. [§III-D] The phrase 'simulation-informed analytical gradients' is useful, but the paper also repeatedly says 'exact gradients'. Please distinguish the exact gradients of the algebraic SP model from the approximate β-scaled gradients of the hybrid model to avoid overclaiming.
  3. [Algorithm 4, line 18-19] The discretization v0_i = I(P_i ≥ 0.5), v1_i = 1 − v0_i forces every startpoint to toggle. This is a reasonable heuristic for maximizing transitions, but it is a design choice not derived from the optimization; please acknowledge this limitation, especially for designs where reconvergence may make complementary primary-input toggling suboptimal.
  4. [§V-B / Table III] The text in §VI says the large design has TR r=0.84, while Table III reports 0.835. Rounding is fine, but please keep the table and text consistent or explicitly state the rounding.
  5. [Abstract] The abstract states 'near-perfect rank agreement' without specifying that the agreement is with finite-difference gradients of the same differentiable model, not with signoff power gradients. A short qualifier would prevent misinterpretation.
  6. [General] The paper does not mention a code/data release. Given the strong runtime and portability claims, a reproducibility artifact (even for the open-source IWLS benchmarks) would substantially increase confidence and is encouraged.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: core accuracy claims are validated against external commercial reference and power-virus evaluation uses a distinct discrete ground-truth metric.

full rationale

DiffPower's central claims do not reduce to their inputs by construction. The toggle-rate accuracy claim (median r=0.96) is evaluated against an external commercial reference (Synopsys Fusion Compiler/Design Compiler) using per-pin correlation, not against DiffPower's own outputs (Table III, Section V-B). The beta-scaling factors (Eq. 5) are computed from bit-packed simulation, not from the commercial reference, and the hybrid TR is then compared to the commercial tool, so this is an external check. The gradient validation in Section V-C verifies that reverse-mode AD matches finite differences of DiffPower's own forward model; this is an implementation-correctness test, not a prediction claim, and the paper states it as such ('produces exact derivatives of the forward SP model'). The power-virus results are evaluated on the true discrete P_virus (Eq. 3) via bitwise simulation, independent of the differentiable proxy (Eq. 9) used for search, so the 2.13x improvement is not a self-referential fit. The cell-sizing experiment may rely on an approximate power model for acceptance/evaluation, but the paper does not define the sizing gain as an identity; it is a method comparison on an optimization objective, and any external-validity limitation is a correctness concern rather than circularity. No load-bearing self-citations or imported uniqueness theorems are present; the straight-through-estimator analogy to prior work is illustrative, not a circular justification. Therefore no circular step can be exhibited under the required standard.

Assumptions & free parameters 6 free parameters · 5 assumptions · 0 invented entities

The central claims rest on standard EDA domain assumptions (temporal independence, combinational scoping, commercial tool as ground truth) plus an ad hoc straight-through estimator that connects the simulation-calibrated forward model to the analytical gradients. The per-pin beta factors are fitted to simulation data, and the clipping bounds and MC count are manually chosen. No genuinely new physical entities are introduced.

free parameters (6)
  • beta_min (clipping lower bound) = 0.1
    Chosen by hand to prevent gradient suppression; setting affects hybrid TR accuracy and gradient scaling.
  • beta_max (clipping upper bound) = 10
    Chosen by hand to prevent gradient explosion.
  • Monte Carlo iterations N_sim = 100
    Number of random 64-bit vectors used to compute TR_acc; higher would reduce noise but increase runtime.
  • gradient ascent learning rate eta = not specified
    Used in Algorithm 4; no value reported, so the power virus results are not reproducible as stated.
  • gradient ascent steps T = 100
    Fixed number of ascent steps; affects runtime and final virus quality.
  • finite-difference epsilon = 1e-4
    Used in gradient validation; standard chosen perturbation.
assumptions (5)
  • domain assumption Temporal independence: TR = 2P(1-P)
    Section II-B: toggle rate derived from static probability under independence of successive states; neglects temporal and (before beta scaling) spatial correlations. It is the basis for all analytical gradients.
  • domain assumption Combinational-boundary scoping: sequential elements are fixed PPIs/PPOs with constant SP
    Section III-A: outputs of FFs treated as pseudo-primary inputs with fixed probabilities; limits accuracy on sequential-heavy designs, acknowledged in Section VI as a limitation (large design TR r=0.84).
  • ad hoc to paper Straight-through estimator: backward pass treats beta_i as a constant and uses the analytical SP gradient as the direction of update
    Section III-D, Eq. (6): gradient is beta_i times the analytical SP gradient, not the true derivative of the hybrid forward model. Paper relies on biased-gradient convergence (cites [27], [28]) but does not verify the bias is small for power optimization.
  • domain assumption Bit-packed simulation with 64-bit Markov chains captures the spatial correlations needed for calibration
    Section III-C: 64-bit vectors generated via Markov chains calibrated to target SP/TR; only 100 iterations, so TR_acc estimates are noisy but assumed adequate for beta calibration.
  • domain assumption Commercial reference tool's propagated switching activity is ground truth for correlation
    Section V-A: accuracy baseline is Synopsys Fusion/DC propagate_switching_activity; the paper assumes this is the correct target for optimization.

how reviews work

0 comments
Cite this review

Pith. "Pith review of DiffPower: GPU-Accelerated Differentiable Switching Power Analysis and Optimization." pith.science (2026). https://pith.science/paper/CR4Y45QM

@misc{pith2026260803778,
  author       = {Pith},
  title        = {Pith review of: DiffPower: GPU-Accelerated Differentiable Switching Power Analysis and Optimization},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/CR4Y45QM}},
  note         = {Machine review of arXiv:2608.03778}
}
abstract

Accurate and scalable switching power analysis remains a critical bottleneck in modern physical design, often forcing a trade-off between computational speed and modeling fidelity. We present DiffPower, a GPU-accelerated framework for differentiable power analysis and optimization. DiffPower translates design netlists into a PDK-agnostic bytecode representation, enabling analytical gradient computation via reverse-mode automatic differentiation, achieving up to a $1{,}002\times$ speedup over single-threaded CPU propagation on the largest evaluated design, with the GPU advantage growing with design scale. A hybrid propagation methodology fusing analytical modeling with parallel simulation achieves a median toggle-rate correlation of $r{=}0.96$ across ten industrial and benchmark designs. The resulting \emph{power gradients}, computed up to $904\times$ faster than CPU finite-difference methods with near-perfect rank agreement, enable two downstream applications: (1) gradient-weighted cell sizing, which achieves up to $2.98\times$ improvement over local-power heuristics on industrial designs, with even stronger advantages at the 117K-cell scale where competing methods plateau; and (2) power virus generation via gradient ascent, which yields up to $2.13\times$ higher transition-weighted power, replacing a search process that traditionally requires hours.

Figures

Figures reproduced from arXiv: 2608.03778 by the authors.

Figure 1
Figure 1. Overview of DiffPower. The framework initializes from a commercial reference tool, builds a levelized pin graph with corresponding [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Per-pin activity correlation between DiffPower and a com [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. Cumulative switching power reduction during iterative cell [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

29 extracted references · 25 canonical work pages

  1. [1]

    Dark silicon and the end of multicore scaling,

    H. Esmaeilzadeh, E. Blem, R. St. Amant, K. Sankaralingam, and D. Burger, “Dark silicon and the end of multicore scaling,” inPro- ceedings of the 38th annual international symposium on Computer architecture, 2011, pp. 365–376

    2011

  2. [2]

    Binary decision diagrams,

    R. E. Bryant, “Binary decision diagrams,” inHandbook of model checking. Springer, 2018, pp. 191–217

    work page 2018

  3. [3]

    INSTA: An ultra-fast, differentiable, statistical static timing analysis engine for industrial physical design applications,

    Y .-C. Lu, Z. Guo, K. Kunal, R. Liang, and H. Ren, “INSTA: An ultra-fast, differentiable, statistical static timing analysis engine for industrial physical design applications,” in2025 62nd ACM/IEEE Design Automation Conference (DAC). IEEE, 2025, pp. 1–7

    2025

  4. [4]

    DiffCCD: Differ- entiable concurrent clock and data optimization,

    Y . Ji, Y . Lu, Z. Zhang, Z. Guo, Y . Lin, and B. Yu, “DiffCCD: Differ- entiable concurrent clock and data optimization,” in2025 IEEE/ACM International Conference on Computer-Aided Design (ICCAD). IEEE, 2025, pp. 1–9

    work page 2025

  5. [5]

    Differentiable physical optimiza- tion,

    Y . Du, Z. Guo, R. Wang, and Y . Lin, “Differentiable physical optimiza- tion,” in2025 IEEE/ACM International Conference on Computer-Aided Design (ICCAD). IEEE, 2025, pp. 1–9

    work page 2025

  6. [6]

    Deep differentiable logic gate networks,

    F. Petersen, C. Borgelt, H. Kuehne, and O. Deussen, “Deep differentiable logic gate networks,” inAdvances in Neural Information Processing Systems, S. Koyejo, S. Mohamed, A. Agarwal, D. Belgrave, K. Cho, and A. Oh, Eds., vol. 35. Curran Associates, Inc., 2022, pp. 2006– 2018

    work page 2022

  7. [7]

    GATSPI: GPU ac- celerated gate-level simulation for power improvement,

    Y . Zhang, H. Ren, A. Sridharan, and B. Khailany, “GATSPI: GPU ac- celerated gate-level simulation for power improvement,” inProceedings of the 59th ACM/EDAC/IEEE Design Automation Conference (DAC). ACM, 2022, pp. 1–6

    work page 2022

  8. [8]

    GRANNITE: Graph neural network inference for transferable power estimation,

    Y . Zhang, H. Ren, and B. Khailany, “GRANNITE: Graph neural network inference for transferable power estimation,” in2020 57th ACM/IEEE Design Automation Conference (DAC). IEEE, 2020, pp. 1–6

    work page 2020

Show all 29 references
  1. [9]

    Reverse-mode automatic differentiation and optimization of GPU kernels via Enzyme,

    W. S. Moses, V . Churavy, L. Paehler, J. H ¨uckelheim, S. K. Narayanan, M. Schanen, and J. Doerfler, “Reverse-mode automatic differentiation and optimization of GPU kernels via Enzyme,” inProceedings of the International Conference for High Performance Computing, Networking, S...

    2021

  2. [10]

    Icarus Verilog,

    S. Williams, “Icarus Verilog,” https://steveicarus.github.io/iverilog/, 1998, open-source Verilog simulator; VVP (Verilog Virtual Processor) compiles HDL to a structural-functor bytecode executed by a virtual machine

    1998

  3. [11]

    Strongly NP-hard discrete gate-sizing problems,

    W. Ning, “Strongly NP-hard discrete gate-sizing problems,”IEEE Trans- actions on Computer-Aided Design of Integrated Circuits and Systems, vol. 13, no. 8, pp. 1045–1051, 1994

    1994

  4. [12]

    Timing and area op- timization for standard-cell VLSI circuit design,

    W. Chuang, S. S. Sapatnekar, and I. N. Hajj, “Timing and area op- timization for standard-cell VLSI circuit design,”IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, vol. 14, no. 3, pp. 308–320, 1995

    1995

  5. [13]

    LEGO-SiZE: LLM-enhanced GPU-optimized signoff-accurate differen- tiable VLSI gate sizing in advanced nodes,

    Y .-C. Lu, K. Kunal, G. Pradipta, R. Liang, R. Gandikota, and H. Ren, “LEGO-SiZE: LLM-enhanced GPU-optimized signoff-accurate differen- tiable VLSI gate sizing in advanced nodes,” inProceedings of the 2025 International Symposium on Physical Design, 2025, pp. 152–162

    2025

  6. [14]

    AGD: Analytic gradient descent for discrete optimization in EDA and its use to gate sizing,

    P. Pham, T.-M. Park, S.-H. Cho, T. Mahmood, J.-S. Yang, and J. Chung, “AGD: Analytic gradient descent for discrete optimization in EDA and its use to gate sizing,”ACM Transactions on Design Automation of Electronic Systems, vol. 30, no. 5, pp. 1–22, 2025

    2025

  7. [15]

    Learning- driven physically aware large-scale circuit gate sizing,

    Y . Ye, P. Xu, L. Ren, T. Chen, H. Yan, B. Yu, and L. Shi, “Learning- driven physically aware large-scale circuit gate sizing,”IEEE Transac- tions on Computer-Aided Design of Integrated Circuits and Systems, vol. 44, no. 5, p. 1901, 2025

    1901

  8. [16]

    DREAM- Place: Deep learning toolkit-enabled GPU acceleration for modern VLSI placement,

    Y . Lin, S. Dhar, W. Li, H. Ren, B. Khailany, and D. Z. Pan, “DREAM- Place: Deep learning toolkit-enabled GPU acceleration for modern VLSI placement,” inProceedings of the 56th Annual Design Automation Conference 2019, 2019, pp. 1–6

    2019

  9. [17]

    DGR: Differentiable global router,

    W. Li, R. Liang, A. Agnesina, H. Yang, C.-T. Ho, A. Rajaram, and H. Ren, “DGR: Differentiable global router,” inProceedings of the 61st ACM/IEEE Design Automation Conference, 2024, pp. 1–6

    2024

  10. [18]

    SAGA: A surrogate-assisted genetic algorithm for fast CPU power virus genera- tion,

    P. Chatzimiltis, G. Antoniou, H. V olos, and Y . Sazeides, “SAGA: A surrogate-assisted genetic algorithm for fast CPU power virus genera- tion,” in2025 IEEE International Symposium on Performance Analysis of Systems and Software (ISPASS). IEEE, 2025, pp. 309–319

    2025

  11. [19]

    Exact and approximate estimation for maximum instantaneous current of CMOS circuits,

    Y .-M. Jiang and K.-T. Cheng, “Exact and approximate estimation for maximum instantaneous current of CMOS circuits,” inProceedings Design, Automation and Test in Europe. IEEE, 1998, pp. 698–702

    1998

  12. [20]

    Maximizing the switching activity of different modules within a processor core via evolutionary techniques,

    N. I. Deligiannis, R. Cantoro, and M. S. Reorda, “Maximizing the switching activity of different modules within a processor core via evolutionary techniques,” in2021 24th Euromicro Conference on Digital System Design (DSD). IEEE, 2021, pp. 535–540

    2021

  13. [21]

    System-level max power (SYMPO): A systematic approach for escalating system-level power consumption using synthetic benchmarks,

    K. Ganesan, J. Jo, W. L. Bircher, D. Kaseridis, Z. Yu, and L. K. John, “System-level max power (SYMPO): A systematic approach for escalating system-level power consumption using synthetic benchmarks,” inProceedings of the 19th international conference on Parallel archi- tectur...

    2010

  14. [22]

    Automating the generation of programs maximizing the sustained switching activity in microprocessor units via evolutionary techniques,

    N. I. Deligiannis, R. Cantoro, and M. S. Reorda, “Automating the generation of programs maximizing the sustained switching activity in microprocessor units via evolutionary techniques,”Microprocessors and Microsystems, vol. 98, p. 104775, 2023

    2023

  15. [23]

    Efficient peak power estima- tion using probabilistic cost-benefit analysis,

    H. Hajimiri, K. Rahmani, and P. Mishra, “Efficient peak power estima- tion using probabilistic cost-benefit analysis,” in2015 28th International Conference on VLSI Design. IEEE, 2015, pp. 369–374

    2015

  16. [24]

    DEFT: Differentiable automatic test pattern generation,

    W. Li, Y . Zou, Y . Liang, J. Moura, and S. Blanton, “DEFT: Differentiable automatic test pattern generation,” 2026. [Online]. Available: https://arxiv.org/abs/2512.23746

    2026 arXiv

  17. [25]

    A survey of power estimation techniques in VLSI circuits,

    F. Najm, “A survey of power estimation techniques in VLSI circuits,” IEEE Transactions on V ery Large Scale Integration (VLSI) Systems, vol. 2, no. 4, pp. 446–455, 1994

    1994

  18. [26]

    Quantized neural networks: Training neural networks with low preci- sion weights and activations,

    I. Hubara, M. Courbariaux, D. Soudry, R. El-Yaniv, and Y . Bengio, “Quantized neural networks: Training neural networks with low preci- sion weights and activations,”Journal of Machine Learning Research, vol. 18, no. 187, pp. 1–30, 2018

    2018

  19. [27]

    Understanding straight-through estimator in training activation quantized neural nets,

    P. Yin, J. Lyu, S. Zhang, S. Osher, Y . Qi, and J. Xin, “Understanding straight-through estimator in training activation quantized neural nets,” arXiv preprint arXiv:1903.05662, 2019

    1903 arXiv

  20. [28]

    AdaSTE: An adaptive straight-through estimator to train binary neural networks,

    H. Le, R. K. Høier, C.-T. Lin, and C. Zach, “AdaSTE: An adaptive straight-through estimator to train binary neural networks,” inProceed- ings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2022, pp. 460–469

    2022

  21. [29]

    IWLS 2005 benchmarks,

    International Workshop on Logic Synthesis, “IWLS 2005 benchmarks,” http://iwls.org/iwls2005/benchmarks.html, 2005, 84 industrial and aca- demic Verilog netlists, re-mapped to target library

    2005

Pith tools

Reviewed August 5, 2026 · model on record in the stance chip above.