Pith. sign in

REVIEW 2 major objections 5 minor 30 references

Logic-gate networks match or beat standard neural nets on EEG classification while running as pure bitwise circuits, achieving up to 2.9x faster inference and 14x smaller models on a low-power edge CPU.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

T0 review · deepseek-v4-flash

2026-08-01 15:49 UTC pith:LKGTVKYM

load-bearing objection First real EEG benchmark for Diff-Logic; the latency scaling is well-demonstrated, but the dementia accuracy claim is confounded by a 15x thermometer input expansion. the 2 major comments →

arxiv 2607.18149 v1 pith:LKGTVKYM submitted 2026-07-20 cs.LG cs.AI

Differentiable Logic Gate Networks for Low-Latency EEG Classification on Edge Devices

Shyamal Y. Dharia , Stephen D. Smith , Camilo E. Valderrama This is my paper
classification cs.LG cs.AI
keywords differentiable logic gate networksEEG classificationedge deploymentbrain-computer interfaceBoolean circuitsthermometer encodinglow-latency inferenceJetson benchmarking
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

The paper tries to establish that differentiable logic gate networks (Diff-Logic) — networks whose weights are Boolean gates rather than floating-point numbers — can classify EEG signals accurately enough for clinical and affective tasks while running as pure bitwise circuits on low-power hardware. In iso-parameter comparisons across four datasets, Diff-Logic beat a float32 MLP by up to 6.8% macro F1 on dementia detection and matched MLP accuracy on emotion recognition at the largest scale, while its inference latency stayed nearly constant across a 10x increase in parameters. On a 7W Jetson Orin Nano CPU, compiled Diff-Logic ran 2.9x faster than the MLP and used 14x less memory. A sympathetic reader would care because this points to a hardware-native path for real-time brain-computer interfaces that need microsecond feedback without cloud offloading.

Core claim

On the paper's own terms: Diff-Logic networks, trained through a differentiable relaxation of all 16 two-input Boolean gates and then hardened to sparse Boolean circuits, match or outperform matched-capacity MLP and BNN baselines on four EEG datasets spanning dementia detection and emotion recognition. The best dementia result is 80.2% macro F1 (CN vs AD), versus 73.4% for the best MLP tier; on SEED emotion recognition the MLP retains a moderate lead until the 500k tier, where Diff-Logic ties it (61.0% vs 61.3%) while running 2.3x faster and using 14x less storage. The authors attribute the efficiency to the compiled circuit's depth-bounded evaluation: latency depends on the number of sequen

What carries the argument

The central object is the soft logic gate: each neuron keeps a probability distribution over the 16 Boolean functions of two inputs, computes a soft expected output during training, and is hardened to the argmax function at the end, yielding a static circuit of AND/OR/XOR/NOT gates. Thermometer encoding converts each continuous EEG feature into 15 bits ordered by magnitude, expanding the 95-dim dementia features to 1,425 binary inputs; a Group Sum layer turns the final layer's bits into class logits. Inference is a C-compiled bitwise pass over packed registers, with no multiply-accumulate and no accumulation registers.

Load-bearing premise

The load-bearing premise is that matching parameter counts makes the comparison fair, even though Diff-Logic receives a 15x wider binary input; if the extra input dimensionality rather than the logic-gate inductive bias explains the accuracy gains, the central performance claim collapses.

What would settle it

Train an MLP with the same parameter budget on the same thermometer-encoded binary features (1,425 dims for dementia) and compare macro F1; if it reaches or exceeds 80.2% on CN vs AD, then the advantage comes from the input encoding, not from logic gates.

Watch this falsifier — get emailed when new claim-graph text bears on it.

If this is right

  • Real-time BCI feedback loops can run on single-core, 7W CPUs at ~0.2 ms per inference, two orders of magnitude below real-time thresholds.
  • Model scale can be increased 10x without raising inference cost, so the performance ceiling for logic networks is set by training, not by deployment latency.
  • Logic-gate networks are accumulation-free, giving them a structural advantage over binary neural networks that must still popcount after XNOR.
  • In clinical settings with small cohorts, the logic inductive bias appears to generalize better than the MLP baseline at identical parameter budgets.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The paper's iso-parameter design holds parameter count fixed but not input dimensionality: MLP/BNN see 95–310 float features while Diff-Logic sees 1,425–4,650 thermometer bits. A control experiment feeding the same binary features to an MLP would separate the effect of the logic inductive bias from the effect of the wider binary representation.
  • If the thermometer-encoding advantage is confirmed, the method could transfer to other structured biosignals (ECG, EMG) where clinician knowledge is naturally threshold-based.
  • The flat-latency scaling suggests that logic circuits could push toward much larger models on edge hardware; the practical limit would be compile time and training memory, not runtime.
  • Compiled Boolean circuits are easily mapped to FPGA/ASIC primitives, so the 0.2 ms CPU figure is likely an upper bound; silicon implementations could go faster.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

2 major / 5 minor

Summary. The paper evaluates Differentiable Logic Gate Networks (Diff-Logic) for EEG classification on edge devices. It compares Diff-Logic against MLP and BNN baselines at matched parameter counts (50k–500k) on two dementia binary tasks (CN vs AD, CN vs FTD) and three SEED emotion 3-class tasks (Chinese, French, German). The authors report that Diff-Logic outperforms MLP on dementia (e.g., 80.2% vs 73.4% Macro F1 at the 100k tier) and is competitive on SEED at the largest scale, while compiled Boolean circuits exhibit near-constant latency and achieve up to a 2.91× speedup and 14× smaller model size compared to ONNX Runtime MLPs on a Jetson Orin Nano single-core CPU.

Significance. If the performance results hold, this is a meaningful demonstration of hardware-native logic networks for resource-constrained BCI applications. The latency benchmarking is careful: using untrained Diff-Logic circuits is justified because compiled latency depends only on gate count and connectivity, and comparing against ONNX Runtime on a single-core ARM CPU is a reasonable embedded inference scenario. The flat scaling of Boolean circuit latency with model size is an architecturally interesting result, and the public code release is a strength. However, the central performance claim is not yet established because the comparison is iso-parameter but not iso-input: Diff-Logic receives a 15× thermometer-expanded binary input representation while MLP and BNN receive the original continuous features. The paper's own limitations section acknowledges this expansion but provides no ablation to quantify its effect. Without a matched-input control, the conclusion that Diff-Logic has a 'superior inductive bias' for EEG is confounded.

major comments (2)
  1. [§3.3.1, §4.1, Table A4, Table 1, §6.1, §6.4] The experimental design does not control input dimensionality. §3.3.1 states that thermometer encoding with T=15 expands the 95-dimensional dementia features to 1,425 binary inputs and the 310-dimensional SEED features to 4,650 binary inputs, while §4.1 confirms that MLP and BNN receive the original 95/310 continuous features. Table A4 shows that the architectures are matched only in parameter count, not in input width or first-layer connectivity. Consequently, the dementia accuracy gap in Table 1 (e.g., 80.2% vs 73.4% at the 100k tier) could be attributable to the richer binary input representation rather than to the logic-gate inductive bias. §6.4 explicitly acknowledges the 15× expansion but no ablation is provided. To support the §6.1 claim of 'superior inductive bias,' the authors should add matched-input controls, such as feeding the same thermometer-encoded bits to the MLP and BNN
  2. [§4.3, Table 2, §5.2] The BNN baseline is labeled 'Binary' in Table 2, but the reported model sizes (e.g., ≈198 KB at 50k parameters) are consistent with float32 storage of the weights (50k × 4 bytes ≈ 200 KB), not with packed binary weights. The ONNX Runtime deployment therefore likely uses standard floating-point operations, making the BNN latency and size numbers not representative of true binary inference (XNOR-popcount). This undermines the §5.2 claim that Diff-Logic is 'fundamentally more hardware-efficient than simple weight quantization' based on comparisons against this BNN. The authors should either provide a genuinely binary-packed deployment for the BNN or explicitly state that the BNN is used only as an accuracy baseline and not as an efficiency baseline.
minor comments (5)
  1. [Abstract vs §6.1/§7] The abstract reports a 6.8% F1 improvement (80.2 vs 73.4), while the conclusion reports a 9.4 percentage-point improvement. These refer to different tiers (100k and 200k, respectively). Please use a consistent tier or explicitly state both tier-specific differences.
  2. [Equation (3.3)] The equation for the soft output is malformed: the division is missing. It should read y_soft = Σ_i (exp(α_{n,i}) / Σ_j exp(α_{n,j})) · f_i(x1, x2).
  3. [Table 1, Appendix D] The 'Params' column mixes approximate and exact values (e.g., '≈50k / 75k'). Since Table A5 gives exact counts, please harmonize the notation to avoid confusion about which tier applies to which dataset.
  4. [Figures 2 and 3] Figures 2 and 3 show latency and F1 values but no error bars or confidence intervals, even though five seeds were used. Adding variability information would strengthen the robustness claims.
  5. [§4.2] Diff-Logic uses a learning rate of 0.01 while MLP/BNN use 0.001. This difference is inherited from prior work, but for a rigorous iso-parameter comparison, a short hyperparameter sweep or an argument that the chosen rates are optimal for each model should be reported.

Circularity Check

0 steps flagged

No circularity found; the results are measured against external datasets and prior-work hyperparameters, with no fitted-to-target derivations or self-citation chain.

full rationale

The paper is an empirical benchmark, not a derivation. The central claims (80.2% vs 73.4% Macro F1 on dementia, 2.91x speedup, and flat latency scaling) are measured outcomes from experiments on public EEG datasets, not quantities derived from the model definitions. Hyperparameters tau=30 and eta=0.01 are explicitly taken from Petersen et al. [7], an external prior-work source by different authors, and T=15 is justified via external empirical studies [27,28]; none of these are tuned to the reported F1 scores or latencies. There is no fitted input called a prediction: Diff-Logic, MLP, and BNN are trained and evaluated under a fixed protocol, and no parameter is recovered from the target metric. The paper contains no self-citations and no imported uniqueness theorem; all cited prior work (Diff-Logic, BNNs, thermometer encoding) is external. The acknowledged 15x thermometer input expansion (§6.4) and the absence of a matched-input ablation is a genuine threat to the interpretation of the performance gap, but it is a confound in the comparison, not a circularity: the binary input representation is not constructed from the labels or from the reported outcomes, and the performance claim does not reduce to an assumption by construction. Therefore, no circular step can be exhibited, and the appropriate score is 0.

Axiom & Free-Parameter Ledger

1 free parameters · 4 axioms · 0 invented entities

The central claims rest on feature adequacy, the thermometer encoding choice, and the standard Diff-Logic assumption that hardened circuits preserve soft training accuracy. No new entities are introduced; the only hand-chosen design knob with direct comparison impact is T=15.

free parameters (1)
  • Thermometer encoding level T = 15
    Hand-chosen based on cited prior work; expands each EEG feature to 15 binary inputs, widening Diff-Logic's input space 15x relative to MLP/BNN. The central performance comparison depends on this choice.
axioms (4)
  • domain assumption PSD and differential entropy features contain sufficient discriminative information for dementia and emotion classification
    Task setup relies on pre-extracted tabular features rather than raw EEG; if these features are insufficient, Diff-Logic performance claims are conditional on them (§3.1).
  • domain assumption Thermometer encoding with T=15 preserves task-relevant signal (negligible accuracy loss vs Float32)
    Invoked in §3.3.1; supported by citation [28], which is about 4-bit quantization, not thermometer encoding. The assumption is load-bearing for the input representation comparison.
  • domain assumption Gate hardening after soft training preserves accuracy
    Standard Diff-Logic assumption (§3.3.1); the paper does not report soft-vs-hardened accuracy, yet reports hardened F1 as the central result. If hardening degrades accuracy, reported scores would not reflect the trained soft network.
  • domain assumption Bitwise circuit latency is independent of the specific Boolean functions assigned
    Used to justify benchmarking untrained circuits (§4.3). Reasonable for bitwise operations, but still a hardware-level assumption.

pith-pipeline@v1.3.0-alltime-deepseek · 12255 in / 11697 out tokens · 101201 ms · 2026-08-01T15:49:08.019004+00:00 · methodology

0 comments
read the original abstract

Real-time EEG classification on edge devices is bottlenecked by the floating-point arithmetic of conventional neural networks. We investigated Differentiable Logic Gate Networks (Diff-Logic) as a hardware-native alternative that compiles models into pure Boolean circuits executable via bitwise CPU operations. Through rigorous iso-parameter experiments across four EEG datasets spanning two classification tasks, binary dementia detection and 3-class emotion recognition, we compared Diff-Logic against matched-capacity Multi-Layer Perceptron (MLP) and Binarized Neural Network (BNN) baselines at four complexity tiers (50k-500k parameters). On dementia screening, Diff-Logic achieved 80.2% Macro F1, outperforming the MLP baseline by 6.8%. On emotion recognition, the MLP retained a moderate performance advantage but incurred a 2.3$\times$ higher latency and 14$\times$ larger model size when deployed on a power-constrained (7W) Nvidia Jetson Orin Nano CPU (Single-core). Critically, Diff-Logic inference time remained nearly constant across a 10$\times$ increase in model scale, achieving a peak speedup of 2.9$\times$ over MLPs at the largest complexity tier. Our results establish logic-based neural architectures as a practical paradigm for resource-constrained brain-computer interfaces, achieving competitive or superior performance while natively satisfying the latency and memory constraints of portable edge deployment. Code is available on GitHub: https://github.com/Shyamal-Dharia/eeg-difflogic

Figures

Figures reproduced from arXiv: 2607.18149 by Camilo E. Valderrama, Shyamal Y. Dharia, Stephen D. Smith.

Figure 1
Figure 1. Figure 1: Proposed workflow for EEG classification using Diff-Logic. paradigm has emerged through the work of Petersen et al. [7], who introduced Differen￾tiable Logic Gate Networks. Unlike BNNs, which quantize a fixed topology, this framework learns the topology itself. By relaxing Boolean logic gates (AND, OR, XOR) into continu￾ous differentiable operators, these networks can be trained via gradient descent to ide… view at source ↗
Figure 2
Figure 2. Figure 2: Inference scaling on the Jetson Orin Nano embedded CPU (single-threaded). (a) Dementia (N vs. AD) (b) SEED (3-class Emotion) [PITH_FULL_IMAGE:figures/full_fig_p009_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Macro F1-Score vs. Inference Latency on the Jetson Orin Nano embedded CPU (single-threaded). Each point represents one model tier (50k–500k parameters) and bubble area is proportional to compiled model storage size in KB. 50k to 500k, indicating overfitting. This highlights again the practical value of logic-based models in clinical settings with smaller cohorts, where they offer superior generalization al… view at source ↗

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

30 extracted references · 5 linked inside Pith

  1. [1]

    Deep Learning Approaches for EEG-Motor Imagery-Based BCIs: Current Models, Generalization Challenges, and Emerging Trends

    A. Raza and M. Z. Yusoff. “Deep Learning Approaches for EEG-Motor Imagery-Based BCIs: Current Models, Generalization Challenges, and Emerging Trends”. In:IEEE Access13 (2025), pp. 151866–151893

    2025

  2. [2]

    A Simple Review of EEG Foundation Models: Datasets, Advancements and Future Perspectives

    J. Lai, J. Wei, L. Yao, and Y. Wang. “A Simple Review of EEG Foundation Models: Datasets, Advancements and Future Perspectives”. In:arXiv preprint arXiv:2504.20069(2025)

    arXiv 2025

  3. [3]

    Challenges, Limitations, and Future Directions in EEG-Driven Image Classification using Machine Learning

    A. Joel. “Challenges, Limitations, and Future Directions in EEG-Driven Image Classification using Machine Learning”. In: (May 2024)

    2024

  4. [4]

    Eeg founda- tion models: A critical review of current progress and future directions

    G. Kuruppu, N. Wagh, V. Kremen, S. Pati, G. Worrell, and Y. Varatharajah. “Eeg founda- tion models: A critical review of current progress and future directions”. In:arXiv preprint arXiv:2507.11783(2025)

    arXiv 2025

  5. [5]

    A Comprehensive Survey on Wearable Computing for Mental and Physical Health Monitoring

    S. Muter, F. Alqurashi, and A. Badr. “A Comprehensive Survey on Wearable Computing for Mental and Physical Health Monitoring”. In:Electronics14.17 (2025), p. 3443

    2025

  6. [6]

    A low-latency neural inference framework for real-time handwriting recogni- tion from EEG signals on an edge device: O. Sen et al

    O. Sen, R. Soni, D. Virmani, A. Parekh, P. Lehman, S. Jena, A. Katikhaneni, A. Khalifa, and B. Chatterjee. “A low-latency neural inference framework for real-time handwriting recogni- tion from EEG signals on an edge device: O. Sen et al.” In:Scientific Reports15.1 (2025), p. 41040

    2025

  7. [7]

    Deep differentiable logic gate networks

    F. Petersen, C. Borgelt, H. Kuehne, and O. Deussen. “Deep differentiable logic gate networks”. In:Advances in Neural Information Processing Systems35 (2022), pp. 2006–2018

    2022

  8. [8]

    Convolutional differentiable logic gate networks

    F. Petersen, H. Kuehne, C. Borgelt, J. Welzel, and S. Ermon. “Convolutional differentiable logic gate networks”. In:Advances in Neural Information Processing Systems37 (2024), pp. 121185–121203

    2024

  9. [9]

    Recurrent deep differentiable logic gate networks

    S. Bührer, A. Plesner, T. Aczel, and R. Wattenhofer. “Recurrent deep differentiable logic gate networks”. In:Proceedings of the 2nd International Workshop on Edge and Mobile Foundation Models. 2025, pp. 31–36

    2025

  10. [10]

    Light Differentiable Logic Gate Networks

    L. Rüttgers, T. Aczel, A. Plesner, and R. Wattenhofer. “Light Differentiable Logic Gate Networks”. In:arXiv preprint arXiv:2510.03250(2025). 12

    arXiv 2025

  11. [11]

    Deep learning with convolutional neural networks for EEG decoding and visualization

    R. T. Schirrmeister, J. T. Springenberg, L. D. J. Fiederer, M. Glasstetter, K. Eggensperger, M. Tangermann, F. Hutter, W. Burgard, and T. Ball. “Deep learning with convolutional neural networks for EEG decoding and visualization”. In:Human brain mapping38.11 (2017), pp. 5391–5420

    2017

  12. [12]

    EEGNet: a compact convolutional neural network for EEG-based brain–computer inter- faces

    V. J. Lawhern, A. J. Solon, N. R. Waytowich, S. M. Gordon, C. P. Hung, and B. J. Lance. “EEGNet: a compact convolutional neural network for EEG-based brain–computer inter- faces”. In:Journal of neural engineering15.5 (2018), p. 056013

    2018

  13. [13]

    EEG conformer: Convolutional transformer for EEG decoding and visualization

    Y. Song, Q. Zheng, B. Liu, and X. Gao. “EEG conformer: Convolutional transformer for EEG decoding and visualization”. In:IEEE Transactions on Neural Systems and Rehabilitation Engineering31 (2022), pp. 710–719

    2022

  14. [14]

    Large brain model for learning generic representa- tions with tremendous EEG data in BCI

    W.-B. Jiang, L.-M. Zhao, and B.-L. Lu. “Large brain model for learning generic representa- tions with tremendous EEG data in BCI”. In:arXiv preprint arXiv:2405.18765(2024)

    Pith/arXiv arXiv 2024

  15. [15]

    Biot: Biosignal transformer for cross-data learning in the wild

    C. Yang, M Westover, and J. Sun. “Biot: Biosignal transformer for cross-data learning in the wild”. In:Advances in Neural Information Processing Systems36 (2023), pp. 78240–78260

    2023

  16. [16]

    DeeperBrain: A Neuro-Grounded EEG Foundation Model Towards Universal BCI

    J. Wang, S. Zhao, Y. Zhou, Y. Kang, S. Li, and G. Pan. “DeeperBrain: A Neuro-Grounded EEG Foundation Model Towards Universal BCI”. In:arXiv preprint arXiv:2601.06134(2026)

    arXiv 2026

  17. [17]

    LUNA: Efficient and topology-agnostic foundation model for EEG signal analysis

    B. Döner, T. M. Ingolfsson, L. Benini, and Y. Li. “LUNA: Efficient and topology-agnostic foundation model for EEG signal analysis”. In:arXiv preprint arXiv:2510.22257(2025)

    arXiv 2025

  18. [18]

    FEMBA: Efficient and Scalable EEGAnalysiswithaBidirectionalMambaFoundationModel

    A. Tegon, T. M. Ingolfsson, X. Wang, L. Benini, and Y. Li. “FEMBA: Efficient and Scalable EEGAnalysiswithaBidirectionalMambaFoundationModel”.In:arXiv preprint: 2502.06438 (2025)

    arXiv 2025

  19. [19]

    Distilling the knowledge in a neural network

    G. Hinton, O. Vinyals, and J. Dean. “Distilling the knowledge in a neural network”. In:arXiv preprint arXiv:1503.02531(2015)

    Pith/arXiv arXiv 2015

  20. [20]

    Deep compression: Compressing deep neural networks with pruning, trained quantization and huffman coding

    S. Han, H. Mao, and W. J. Dally. “Deep compression: Compressing deep neural networks with pruning, trained quantization and huffman coding”. In:arXiv preprint arXiv:1510.00149 (2015)

    Pith/arXiv arXiv 2015

  21. [21]

    Binarized neural net- works: Training deep neural networks with weights and activations constrained to+ 1 or-1

    M. Courbariaux, I. Hubara, D. Soudry, R. El-Yaniv, and Y. Bengio. “Binarized neural net- works: Training deep neural networks with weights and activations constrained to+ 1 or-1”. In:arXiv preprint arXiv:1602.02830(2016)

    Pith/arXiv arXiv 2016

  22. [22]

    MuBiNN: Multi-level bina- rized recurrent neural network for EEG signal classification

    S. A. Mirsalari, S. Sinaei, M. E. Salehi, and M. Daneshtalab. “MuBiNN: Multi-level bina- rized recurrent neural network for EEG signal classification”. In:2020 IEEE international symposium on circuits and systems (ISCAS). IEEE. 2020, pp. 1–5

    2020

  23. [23]

    A dataset of scalp EEG recordings of Alzheimer’s disease, frontotemporal dementia and healthy subjects from routine EEG

    A. Miltiadous, K. D. Tzimourta, T. Afrantou, P. Ioannidis, N. Grigoriadis, D. G. Tsalikakis, P. Angelidis, M. G. Tsipouras, E. Glavas, N. Giannakeas, et al. “A dataset of scalp EEG recordings of Alzheimer’s disease, frontotemporal dementia and healthy subjects from routine EEG”. In:Data8.6 (2023), p. 95

    2023

  24. [24]

    Investigating Critical Frequency Bands and Channels for EEG- based Emotion Recognition with Deep Neural Networks

    W.-L. Zheng and B.-L. Lu. “Investigating Critical Frequency Bands and Channels for EEG- based Emotion Recognition with Deep Neural Networks”. In:IEEE Transactions on Au- tonomous Mental Development7.3 (2015), pp. 162–175

    2015

  25. [25]

    Identifying similarities and differences in emotion recognition with EEG and eye movements among Chinese, German, and French People

    W. Liu, W.-L. Zheng, Z. Li, S.-Y. Wu, L. Gan, and B.-L. Lu. “Identifying similarities and differences in emotion recognition with EEG and eye movements among Chinese, German, and French People”. In:Journal of Neural Engineering19.2 (2022), p. 026012

    2022

  26. [26]

    A review of classification algorithms for EEG-based brain–computer interfaces: a 10 year update

    F. Lotte, L. Bougrain, A. Cichocki, M. Clerc, M. Congedo, A. Rakotomamonjy, and F. Yger. “A review of classification algorithms for EEG-based brain–computer interfaces: a 10 year update”. In:Journal of Neural Engineering15.3 (2018), p. 031005

    2018

  27. [27]

    Thermometer encoding: One hot way to resist adversarial examples

    J. Buckman, A. Roy, C. Raffel, and I. Goodfellow. “Thermometer encoding: One hot way to resist adversarial examples”. In:International conference on learning representations. 2018

    2018

  28. [28]

    Post training 4-bit quantization of convolutional networks for rapid-deployment

    R. Banner, Y. Nahshan, and D. Soudry. “Post training 4-bit quantization of convolutional networks for rapid-deployment”. In:Advances in neural information processing systems32 (2019)

    2019

  29. [29]

    3575392- r2

    NVIDIA Corporation.NVIDIA Jetson Orin Nano Super Developer Kit Datasheet. 3575392- r2. Accessed: 2025-02-16. NVIDIA Corporation. Dec. 2024.url:https://nvdam.widen.net/ s/zkfqjmtds2/jetson-orin-datasheet-nano-developer-kit-3575392-r2

    2025

  30. [30]

    Device JNEEG to convert Jetson Nano to brain-Computer interfaces. Short report

    I. Rakhmatulin. “Device JNEEG to convert Jetson Nano to brain-Computer interfaces. Short report”. In:arXiv preprint arXiv:2301.11110(2023). 13 AppendixA.Dataset Summary Table A1.Summary of EEG datasets. Features denote Power Spectral Density (PSD) or Differential Entropy (DE). Windows denotes total non-overlapping segments across all subjects. Dataset Typ...

    Pith/arXiv arXiv 2023