Pith. sign in

REVIEW 4 major objections 6 minor 1 cited by

Sparse Convolutional Recurrent Learning for Efficient Event-based Neuromorphic Object Detection

T0 review · 4 major / 6 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read Sparse convolutional recurrent learning lets event-based object detectors cut synaptic operations by 2x to 6x against state-of-the-art while keeping or improving mean average precision.

desk verdict SEED is a worthwhile, honest extension of sparse recurrent learning to convolutional event-camera detection, with solid mAP and GSOp comparisons, but the hardware latency headline depends on an average-density simulator assumption that should be stress-tested before reviewers sign off. read the letter →

arxiv 2506.13440 v1 pith:APTHNTK2 submitted 2025-06-16 cs.CV cs.NE

Shenqi Wang , Yingfu Xu , Amirreza Yousefzadeh , Sherif Eissa , Henk Corporaal , Federico Corradi , Guangzhi Tang This is my paper · ORCID
classification cs.CVcs.NE
keywords event-basedvisionobjectdetectionsparserecurrentlearningactivationsparsityneuromorphiccomputingeventcameraSENECAconvolutionalunit
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

SEED is a fully convolutional event-camera object detector whose recurrent layers are made sparse enough for low-power neuromorphic hardware. The paper claims that by extending EGRU-style thresholded recurrent processing to convolutional gated units and adding an L1 activation-sparsity loss, recurrent processing runs at over 92% activation sparsity. On the 1Mpx and Gen1 automotive event datasets, SEED reports higher or equal mean average precision than the RED and RVT baselines while needing only a fraction of the synaptic operations (e.g., 3.83 GSOp versus 26.1 for RED on 1Mpx, with mAP 44.9 versus 43.0). Simulated on the SENECA neuromorphic processor, the hardware-aware design translates the sparsity into large energy and latency reductions. If correct, this makes accurate event-camera detection viable on resource-constrained edge devices.

What carries the argument

The load-bearing mechanism is sparse convolutional recurrent (Conv-Rec) processing, a generalization of the EGRU's event-based gating to convolutional gated recurrent units. The update equation $h(t)=\alpha(t)\odot h(t-1)+(1-\alpha(t))\odot z(t)-s(t-1)\odot V_{\mathrm{th}}$ applies a soft reset inspired by spiking neurons, and the Heaviside event generation $s(t)=H(h(t)-V_{\mathrm{th}})$ with sparse hidden output $y(t)=h(t)\odot s(t)$ ensures that all gating convolutions see only sparse activations. A per-neuron learnable threshold $V_{\mathrm{th}}$, trained with surrogate gradients, lets the network choose its own sparsity level; an L1 activation-sparsity loss on the residual blocks extends sparsity to the feedforward part. On SENECA, event-based depth-first convolution skips zero activations, so this sparsity is what converts algorithmic savings into measured reductions in synaptic operations, energy, and latency.

What would settle it

Run SEED-128 and RED on a real SENECA chip (or a cycle-accurate RTL model) with identical 1Mpx event streams and measure per-inference energy and end-to-end latency; if SEED does not show a several-fold advantage over RED and does not land near the simulated 39.3 mJ and 21.6 ms, the hardware-efficiency claim is contradicted. Independently, recompute RED and RVT GSOp with SEED's zero-skipping methodology from the public implementations; if the ratios 26.1-to-3.83 and 8.69-to-2.75 do not reproduce, the algorithmic efficiency claim collapses.

Watch

Extended reading notes

Core claim

The central claim is that convolutional recurrent units—normally the computational bottleneck in event-based detection—can be made event-driven without sacrificing the long-term temporal memory the task needs. In SEED, the hidden state updates as $h(t)=\alpha(t)\odot h(t-1)+(1-\alpha(t))\odot z(t)-s(t-1)\odot V_{\mathrm{th}}$, binary events are generated by $s(t)=H(h(t)-V_{\mathrm{th}})$, and the sparse hidden output $y(t)=h(t)\odot s(t)$ is the only recurrent input to the convolutions; the learnable per-neuron threshold $V_{\mathrm{th}}$ is trained with surrogate gradients. A sparsity loss on the residual-block activations pushes feedforward sparsity as well. On Prophesee's 1Mpx and Gen1 benchmarks, SEED-256 reaches 44.9 and 45.3 mAP with 3.83 and 1.32 GSOp, respectively, improving on RED (43.0/40.0 mAP at 26.1/8.26 GSOp) and matching RVT-S while cutting its GSOp by more than half. The paper also shows that the recurrent module, not the feedforward backbone, is the main source of hardware savings: removing it and the sparsity loss increases simulated energy and latency about fivefold on SENECA.

Load-bearing premise

The efficiency claims rest on two linked assumptions: that the zero-skipping GSOp model counts synaptic operations consistently across SEED, RED, RVT, and the SNN baselines, and that the SENECA micro-kernel simulator, calibrated on earlier chip measurements, predicts real silicon energy and latency; if either assumption fails, the 2x-6x savings and the energy/latency conclusions would be overstated.

Editorial extensions

If this is right

  • Event-camera object detection can be moved from compute-heavy ConvLSTM or transformer backbones to fully convolutional sparse recurrent networks without a mAP penalty.
  • A single algorithmic recipe—thresholded recurrent hidden states plus an activation-sparsity loss—works across GRU, MGU, MinimalRNN, and LSTM designs, with all recurrent layers staying under about 7% event density.
  • Recurrent processing remains necessary: removing it drops group-wise mAP most sharply for objects with little instant event information, so future efficiency work should sparsify temporal memory rather than remove it.
  • On SENECA, shrinking the recurrent channel dimension (SEED-128 versus SEED-256) gives a 2x latency reduction, larger than the raw GSOp reduction, because lower memory pressure lets the mapper parallelize high-latency layers across more cores.
  • Sparsity-aware finetuning reduces synaptic operations by 29-43% with no loss in mAP, indicating accuracy and efficiency do not have to be traded off at this operating point.

Reading between the lines

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

  • Beyond the paper's benchmarks, the group-wise mAP result suggests SEED's advantage will grow as event time bins shrink; testing at sub-millisecond bin widths would directly probe this.
  • Because the sparse recurrent update is unit-agnostic, the same thresholding recipe could be transferred to other spatio-temporal architectures such as video transformers, making them candidates for neuromorphic data-flow execution.
  • The SENECA simulation is instruction-level; an independent silicon measurement would be a stronger test of whether the reported 39.3 mJ and 21.6 ms for SEED-128 hold on real hardware, including any memory or communication overhead the micro-kernels omit.
  • If the GSOp counting methodology were applied to SEED's SSD head and event preprocessing rather than only the recurrent backbone, the remaining dense operations could become the next target for sparsification.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 6 minor

Summary. The paper proposes SEED, a fully convolutional event-based object detector that integrates sparse convolutional recurrent (Conv-Rec) layers inspired by EGRU, together with an activation sparsity loss on residual blocks. The method is evaluated on the Prophesee 1Mpx and Gen1 event-based object detection datasets. The main reported results are: (i) SEED-256 achieves 44.9 mAP on 1Mpx and 45.3 on Gen1 with GSOp of 3.83 and 1.32, respectively, outperforming RED and roughly matching RVT-S with lower GSOp; (ii) sparsity-aware training reduces GSOp by 29-43% without mAP loss; (iii) a SENECA hardware simulation reports energy and latency reductions relative to ablations and a RED variant. The paper frames SEED as a new benchmark in computational efficiency for event-based detection with long-term temporal learning, positioned for deployment on neuromorphic processors.

Significance. If the results hold, the paper makes a useful contribution by showing that convolutional recurrent units can be made highly sparse (over 92% activation sparsity) while retaining the temporal reasoning needed for event-based detection, and by demonstrating that the resulting network can be mapped to a digital neuromorphic processor. The extension of EGRU-style event-based recurrent processing to ConvGRU and its generalization to other recurrent units (MGU, MinimalRNN, LSTM) is a valuable algorithmic step. The authors provide open-source code and compare against several baselines. However, the hardware efficiency claims are not fully validated because the SENECA simulation relies on average activation density and on the authors' own micro-kernel measurements, and several statistical and internal-consistency issues remain. With these addressed, the paper could be a solid addition to the event-based vision and neuromorphic computing literature.

major comments (4)
  1. [Section V.A] The authors state that 'we determine the average required computation of a layer by the average activation density of its inputs over the test data of the 1Mpx dataset.' This averaging is used for both mapping and benchmarking, but event-driven processing latency on SENECA depends on the instantaneous number and spatial distribution of active events per timestep, not the test-set average. Using the mean density removes temporal burstiness and can substantially understate tail latency, especially for recurrent layers whose inputs depend on the previous hidden state. Because Table V's latency and energy results (including the '2x latency reduction' and 'over 80% contribution of the Conv-Rec unit') are all derived from this average-density assumption, the low-latency conclusions are not yet supported. Please provide per-timestep latency distributions (e.g., mean, median, 90th percentile), report the sensitivity of the results to the density estimate, and justify the averaging assumption with evidence that per-timestep variation is small.
  2. [Section V.B / Table V] The simulated 'RED (w/o SE)' is described as a data-flow-friendly version of RED with Squeeze-and-Excitation blocks replaced by sparse residual blocks, yet its reported GSOp (26.1) is identical to the original RED in Table I. If the network is modified, its GSOp must be recomputed; reporting an unchanged value is internally inconsistent and undermines the comparison with SEED. Additionally, the comparison showing RED reaching 40.7 ms latency with 1446 cores versus SEED-256's 44.9 ms with 254 cores should be discussed in terms of the area-latency trade-off; the statement that SEED is 'significantly cheaper' needs to weigh the latency/cores ratio carefully.
  3. [Section IV.B / Table I] The claim that SEED delivers 'higher or same-level mAP' relative to RVT-S is only partly supported by the data: on Gen1, SEED-256 achieves 45.3 mAP versus RVT-S's 46.5, a 1.2-point deficit. Since all mAP and GSOp values come from single runs with no error bars or significance testing, it is impossible to determine whether this difference is meaningful or whether the reported efficiency gains are stable across seeds. Please report mean and standard deviation over at least 3 runs (or a paired analysis) for the main comparisons, or at least specify the number of runs and seed sensitivity.
  4. [Section V.A] The SENECA simulator is calibrated on the authors' own prior hardware measurements [29] and the paper does not validate the simulated energy/latency numbers against actual SENECA measurements for the specific networks considered. Given that the efficiency conclusions rest entirely on this simulation, the paper should either provide a validation experiment (e.g., run a small network on SENECA and compare to simulation) or explicitly state the simulation's limitations and an error model. Without this, the 'energy-efficient and low-latency' hardware claim is not independently established.
minor comments (6)
  1. [Table IV] The row entries such as '-1.86(-33%) +0 -0.68(-34%) +0.9' are ambiguous because they mix GSOp changes and mAP changes without clear column separators; please split into separate columns per dataset with headers.
  2. [Section III.C, Eq. (5)] The sparsity loss is defined over all layers l=1..L, but the text says it applies only to selected activation maps in the first convolutional layer and sparse residual blocks; please clarify the exact set of layers included.
  3. [Section IV.C] The phrase 'less than 7% event density generated by recurrent layers' should define 'event density' precisely (e.g., fraction of recurrent hidden units above threshold averaged over timesteps).
  4. [Figure 2] The group-wise mAP plot would benefit from error bars (or at least number of ground-truth boxes per group) to support the claim of improved detection at low event counts.
  5. [Section IV.A] The statement that a 'thorough hyperparameter search' for beta_sparse was performed is not backed by details; please report the search range, criterion, and why 0.04 was selected.
  6. [Section II.A] The discussion of event encoding methods [12] is brief; clarifying the distinction between per-frame encoding and temporal recurrent processing would help motivate the design.

Circularity Check

0 steps flagged · score 0.0 of 10

No circular derivation: SEED's efficiency numbers are measured from trained networks, and the hardware simulation is calibrated on prior chip measurements rather than constructed to reproduce the claimed result.

full rationale

The paper's central claims are not circular. The activation sparsity of SEED is a measured property of the trained network: the sparsity loss in Eq. (5) is a regularizer applied during finetuning, and the reported GSOp counts follow directly from the measured per-layer activation densities on the test set. These are measurements, not fitted parameters renamed as predictions. The comparisons against RED, RVT, and SNN baselines use open-sourced implementations and the same operation-counting convention applied consistently to the baselines described in Section IV.B. The hardware simulation in Section V uses SENECA as the reference architecture and instruction energy costs obtained from actual hardware measurements in prior work [29], with mapping and depth-first convolution rules from [30]; although these are prior works by the same research group, they are empirical calibration and architectural definitions, not a model constructed to force SEED to win. The use of average activation density for latency estimation in Section V.A is a simplifying modeling assumption and could be challenged on correctness grounds, but it is not a circular step: the latency figure is not defined in terms of the conclusion, and the averaging does not analytically guarantee the reported 2x-6x improvements. No equation in the paper reduces to its own input, no fitted parameter is relabeled as a prediction, and no uniqueness theorem or load-bearing self-citation is invoked to forbid alternative designs. Thus, while the hardware-efficiency conclusion depends on assumptions that deserve scrutiny, the derivation chain itself is self-contained and not circular.

Assumptions & free parameters 3 free parameters · 4 assumptions · 0 invented entities

SEED is an empirical engineering paper, not a parameter-free derivation. The central sparsity-efficiency tradeoff is governed by beta_sparse and architecture width, and the hardware-efficiency claims inherit the assumptions of the SENECA simulation. No new physical entities are postulated.

free parameters (3)
  • beta_sparse = 0.04
    Sparsity loss coefficient selected by 'thorough hyperparameter search' (Section IV.A) to balance mAP and GSOp; it directly controls the sparsity-efficiency tradeoff that the paper's efficiency claims depend on.
  • recurrent channel width = 128 and 256
    SEED-128 and SEED-256 are two model variants in Table I; the architecture choice affects both mAP and GSOp, so the reported efficiency benchmark is partly a product of this choice.
  • bounding box filtering thresholds = 1Mpx: diagonal <60 px, side <20 px; Gen1: diagonal <30 px, side <10 px
    Dataset-specific evaluation heuristics inherited from RED [2]; if applied inconsistently across baselines, the mAP comparisons would shift. They are hand-chosen rules that affect the reported metric.
assumptions (4)
  • domain assumption Surrogate gradient through the Heaviside step function yields a usable training signal for thresholded recurrent layers.
    Invoked in Section III.B using the same methods as EGRU [9]; training relies on an approximate gradient of the step function, which is not proven to yield the reported optima.
  • domain assumption SENECA micro-kernel energy and latency measurements from prior work [29,30] are accurate for SEED's operations.
    Section V.A builds the hardware simulation on these external-measurement inputs; if the measurements or the depth-first convolution model are unrepresentative, simulated energy and latency numbers are not predictive of real chips.
  • domain assumption Average activation density over the 1Mpx test set is representative of deployment conditions.
    Section V.A uses average density to set computational load; dense scenes or distribution shift could reduce realized sparsity and erode the reported efficiency gains.
  • standard math Standard stochastic training (Adam, OneCycle scheduler, SSD loss) converges to the reported models.
    Section IV.A describes a conventional two-stage training pipeline; no formal convergence proof is given, but this is standard practice for deep detection networks.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Sparse Convolutional Recurrent Learning for Efficient Event-based Neuromorphic Object Detection." pith.science (2026). https://pith.science/paper/APTHNTK2

@misc{pith2026250613440,
  author       = {Pith},
  title        = {Pith review of: Sparse Convolutional Recurrent Learning for Efficient Event-based Neuromorphic Object Detection},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/APTHNTK2}},
  note         = {Machine review of arXiv:2506.13440}
}
read the original abstract

Leveraging the high temporal resolution and dynamic range, object detection with event cameras can enhance the performance and safety of automotive and robotics applications in real-world scenarios. However, processing sparse event data requires compute-intensive convolutional recurrent units, complicating their integration into resource-constrained edge applications. Here, we propose the Sparse Event-based Efficient Detector (SEED) for efficient event-based object detection on neuromorphic processors. We introduce sparse convolutional recurrent learning, which achieves over 92% activation sparsity in recurrent processing, vastly reducing the cost for spatiotemporal reasoning on sparse event data. We validated our method on Prophesee's 1 Mpx and Gen1 event-based object detection datasets. Notably, SEED sets a new benchmark in computational efficiency for event-based object detection which requires long-term temporal learning. Compared to state-of-the-art methods, SEED significantly reduces synaptic operations while delivering higher or same-level mAP. Our hardware simulations showcase the critical role of SEED's hardware-aware design in achieving energy-efficient and low-latency neuromorphic processing.

Figures

Figures reproduced from arXiv: 2506.13440 by the authors.

Figure 1
Figure 1. Sparse event-driven efficient detector (SEED) algorithm for event-based object detection. The event-based convolution module receives preprocessed [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Comparing the object detection performance of SEED-256 and its recurrent-free ablation on the 1Mpx dataset. Left - Overall and group-wise mAP [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. Hardware simulation mapping of SEED-256 on a 256-core SENECA neuromorphic processor. Each SENECA core contains an on-chip memory, [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗

Review history (2 revisions) →

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Context-aware Sparse Spatiotemporal Learning for Event-based Vision

    cs.CV 2025-08 conditional novelty 6.0 of 10

    A learned input-adaptive threshold gates neuron activations in event-based neural networks, delivering near-SOTA object detection and optical flow with 32-68% fewer synaptic operations.

Reference graph

Works this paper leans on

33 extracted references · 15 canonical work pages · cited by 1 Pith paper

  1. [29]

    Open the box of digital neuromorphic processor: Towards effective algorithm-hardware co-design,

    G. Tang, A. Safa, K. Shidqi, P. Detterer, S. Traferro, M. Konijnenburg, M. Sifalakis, G.-J. van Schaik, and A. Yousefzadeh, “Open the box of digital neuromorphic processor: Towards effective algorithm-hardware co-design,” in 2023 IEEE International Symposium on Circuits and Systems (ISCAS), 2023, pp. 1–5

    work page 2023

  2. [1]

    Event- based vision: A survey,

    G. Gallego, T. Delbr ¨uck, G. Orchard, C. Bartolozzi, B. Taba, A. Censi, S. Leutenegger, A. J. Davison, J. Conradt, K. Daniilidis et al., “Event- based vision: A survey,” IEEE transactions on pattern analysis and machine intelligence, vol. 44, no. 1, pp. 154–180, 2020

    2020

  3. [2]

    Learning to detect objects with a 1 megapixel event camera,

    E. Perot, P. De Tournemire, D. Nitti, J. Masci, and A. Sironi, “Learning to detect objects with a 1 megapixel event camera,” Advances in Neural Information Processing Systems , vol. 33, pp. 16 639–16 652, 2020

    work page 2020

  4. [3]

    A large scale event-based detection dataset for automotive,

    P. De Tournemire, D. Nitti, E. Perot, D. Migliore, and A. Sironi, “A large scale event-based detection dataset for automotive,” arXiv preprint arXiv:2001.08499, 2020

    arXiv 2001

  5. [4]

    Asynchronous spatio-temporal memory network for continuous event-based object detection,

    J. Li, J. Li, L. Zhu, X. Xiang, T. Huang, and Y . Tian, “Asynchronous spatio-temporal memory network for continuous event-based object detection,” IEEE Transactions on Image Processing , vol. 31, pp. 2975– 2987, 2022

    2022

  6. [5]

    Recurrent vision transformers for object detection with event cameras,

    M. Gehrig and D. Scaramuzza, “Recurrent vision transformers for object detection with event cameras,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , 2023, pp. 13 884–13 893

    2023

  7. [6]

    Event-based optical flow on neuromorphic processor: Ann vs. snn comparison based on activation sparsification,

    Y . Xu, G. Tang, A. Yousefzadeh, G. C. de Croon, and M. Sifalakis, “Event-based optical flow on neuromorphic processor: Ann vs. snn comparison based on activation sparsification,” Neural Networks , p. 107447, 2025

    work page 2025

  8. [7]

    Seneca: building a fully digital neuromorphic processor, design trade-offs and challenges,

    G. Tang, K. Vadivel, Y . Xu, R. Bilgic, K. Shidqi, P. Detterer, S. Traferro, M. Konijnenburg, M. Sifalakis, G.-J. van Schaik et al., “Seneca: building a fully digital neuromorphic processor, design trade-offs and challenges,” Frontiers in Neuroscience, vol. 17, 2023

    work page 2023

Show all 33 references
  1. [8]

    Object detection with spiking neural networks on automotive event data,

    L. Cordone, B. Miramond, and P. Thierion, “Object detection with spiking neural networks on automotive event data,” in2022 International Joint Conference on Neural Networks (IJCNN) . IEEE, 2022, pp. 1–8

    2022

  2. [9]

    Ef- ficient recurrent architectures through activity sparsity and sparse back- propagation through time,

    A. Subramoney, K. K. Nazeer, M. Sch ¨one, C. Mayr, and D. Kappel, “Ef- ficient recurrent architectures through activity sparsity and sparse back- propagation through time,” in The Eleventh International Conference on Learning Representations, 2022

    2022

  3. [10]

    Memory- efficient deep learning on a spinnaker 2 prototype,

    C. Liu, G. Bellec, B. V ogginger, D. Kappel, J. Partzsch, F. Neum ¨arker, S. H ¨oppner, W. Maass, S. B. Furber, R. Legenstein et al. , “Memory- efficient deep learning on a spinnaker 2 prototype,” Frontiers in neuro- science, vol. 12, p. 840, 2018

    2018

  4. [11]

    How many events make an object? improving single-frame object detection on the 1 mpx dataset,

    A. Kugele, T. Pfeil, M. Pfeiffer, and E. Chicca, “How many events make an object? improving single-frame object detection on the 1 mpx dataset,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , 2023, pp. 3912–3921

    2023

  5. [12]

    From chaos comes order: Ordering event representations for object recognition and detection,

    N. Zubi ´c, D. Gehrig, M. Gehrig, and D. Scaramuzza, “From chaos comes order: Ordering event representations for object recognition and detection,” in Proceedings of the IEEE/CVF International Conference on Computer Vision , 2023, pp. 12 846–12 856

    2023

  6. [13]

    Accelerating convolutional neural networks via activa- tion map compression,

    G. Georgiadis, “Accelerating convolutional neural networks via activa- tion map compression,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , 2019, pp. 7085–7095

    2019

  7. [14]

    Inducing and exploiting activation sparsity for fast inference on deep neural networks,

    M. Kurtz, J. Kopinsky, R. Gelashvili, A. Matveev, J. Carr, M. Goin, W. Leiserson, S. Moore, N. Shavit, and D. Alistarh, “Inducing and exploiting activation sparsity for fast inference on deep neural networks,” in International Conference on Machine Learning . PMLR, 2020, pp. 5533–5543

    2020

  8. [15]

    Learning phrase representations using rnn encoder-decoder for statistical machine translation,

    K. Cho, B. van Merrienboer, C. Gulcehre, D. Bahdanau, F. Bougares, H. Schwenk, and Y . Bengio, “Learning phrase representations using rnn encoder-decoder for statistical machine translation,” 2014

    2014

  9. [16]

    Loihi: A neuromorphic manycore processor with on-chip learning,

    M. Davies, N. Srinivasa, T.-H. Lin, G. Chinya, Y . Cao, S. H. Choday, G. Dimou, P. Joshi, N. Imam, S. Jain et al. , “Loihi: A neuromorphic manycore processor with on-chip learning,” Ieee Micro, vol. 38, no. 1, pp. 82–99, 2018

    2018

  10. [17]

    Truenorth: Design and tool flow of a 65 mw 1 million neuron programmable neurosynaptic chip,

    F. Akopyan, J. Sawada, A. Cassidy, R. Alvarez-Icaza, J. Arthur, P. Merolla, N. Imam, Y . Nakamura, P. Datta, G.-J. Nam et al. , “Truenorth: Design and tool flow of a 65 mw 1 million neuron programmable neurosynaptic chip,” IEEE transactions on computer- aided design of integra...

    2015

  11. [18]

    Nxtf: An api and compiler for deep spiking neural networks on intel loihi,

    B. Rueckauer, C. Bybee, R. Goettsche, Y . Singh, J. Mishra, and A. Wild, “Nxtf: An api and compiler for deep spiking neural networks on intel loihi,” ACM Journal on Emerging Technologies in Computing Systems (JETC), vol. 18, no. 3, pp. 1–22, 2022

    2022

  12. [19]

    Towards artificial general intelligence with hybrid tianjic chip architecture,

    J. Pei, L. Deng, S. Song, M. Zhao, Y . Zhang, S. Wu, G. Wang, Z. Zou, Z. Wu, W. He et al., “Towards artificial general intelligence with hybrid tianjic chip architecture,” Nature, vol. 572, no. 7767, pp. 106–111, 2019

    2019

  13. [20]

    Ssd: Single shot multibox detector,

    W. Liu, D. Anguelov, D. Erhan, C. Szegedy, S. Reed, C.-Y . Fu, and A. C. Berg, “Ssd: Single shot multibox detector,” in Computer Vision– ECCV 2016: 14th European Conference, Amsterdam, The Netherlands, October 11–14, 2016, Proceedings, Part I 14 . Springer, 2016, pp. 21–37

    2016

  14. [21]

    Hats: Histograms of averaged time surfaces for robust event-based ob- ject classification,

    A. Sironi, M. Brambilla, N. Bourdis, X. Lagorce, and R. Benosman, “Hats: Histograms of averaged time surfaces for robust event-based ob- ject classification,” in Proceedings of the IEEE conference on computer vision and pattern recognition , 2018, pp. 1731–1740

    2018

  15. [22]

    High speed and high dynamic range video with an event camera,

    H. Rebecq, R. Ranftl, V . Koltun, and D. Scaramuzza, “High speed and high dynamic range video with an event camera,” IEEE transactions on pattern analysis and machine intelligence, vol. 43, no. 6, pp. 1964–1980, 2019

    1964

  16. [23]

    Long Short-Term Memory,

    S. Hochreiter and J. Schmidhuber, “Long Short-Term Memory,” Neural Computation, vol. 9, no. 8, pp. 1735–1780, 11 1997. [Online]. Available: https://doi.org/10.1162/neco.1997.9.8.1735

    1997 doi

  17. [24]

    Minimal gated unit for recurrent neural networks,

    G.-B. Zhou, J. Wu, C.-L. Zhang, and Z.-H. Zhou, “Minimal gated unit for recurrent neural networks,” International Journal of Automation and Computing, vol. 13, no. 3, pp. 226–234, 2016

    2016

  18. [25]

    Minimalrnn: Toward more interpretable and trainable recur- rent neural networks,

    M. Chen, “Minimalrnn: Toward more interpretable and trainable recur- rent neural networks,” arXiv preprint arXiv:1711.06788 , 2017

    2017 arXiv

  19. [26]

    Reducing information loss for spiking neural networks,

    Y . Guo, Y . Chen, L. Zhang, Y . Wang, X. Liu, X. Tong, Y . Ou, X. Huang, and Z. Ma, “Reducing information loss for spiking neural networks,” in Computer Vision – ECCV 2022, S. Avidan, G. Brostow, M. Ciss´e, G. M. Farinella, and T. Hassner, Eds. Cham: Springer Nature Switzerla...

    2022

  20. [27]

    Integer-valued training and spike-driven inference spiking neural network for high-performance and energy-efficient object detection,

    X. Luo, M. Yao, Y . Chou, B. Xu, and G. Li, “Integer-valued training and spike-driven inference spiking neural network for high-performance and energy-efficient object detection,” inComputer Vision – ECCV 2024, A. Leonardis, E. Ricci, S. Roth, O. Russakovsky, T. Sattler, and G...

    2024

  21. [28]

    Microsoft coco: Common objects in context,

    T.-Y . Lin, M. Maire, S. Belongie, J. Hays, P. Perona, D. Ramanan, P. Doll ´ar, and C. L. Zitnick, “Microsoft coco: Common objects in context,” in Computer Vision–ECCV 2014: 13th European Conference, Zurich, Switzerland, September 6-12, 2014, Proceedings, Part V 13 . Springer,...

    2014

  22. [30]

    Optimizing event-based neural networks on digital neuromorphic architecture: a comprehensive design space exploration,

    Y . Xu, K. Shidqi, G.-J. van Schaik, R. Bilgic, A. Dobrita, S. Wang, R. Meijer, P. Nembhani, C. Arjmand, P. Martinello et al., “Optimizing event-based neural networks on digital neuromorphic architecture: a comprehensive design space exploration,” Frontiers in Neuroscience , v...

    2024

  23. [31]

    Squeeze-and-excitation networks,

    J. Hu, L. Shen, and G. Sun, “Squeeze-and-excitation networks,” in Proceedings of the IEEE conference on computer vision and pattern recognition, 2018, pp. 7132–7141

    2018

  24. [32]

    Opportunities for neuromorphic computing algorithms and applica- tions,

    C. D. Schuman, S. R. Kulkarni, M. Parsa, J. P. Mitchell, B. Kay et al., “Opportunities for neuromorphic computing algorithms and applica- tions,” Nature Computational Science , vol. 2, no. 1, pp. 10–19, 2022

    2022

  25. [33]

    Better and faster: Adap- tive event conversion for event-based object detection,

    Y . Peng, Y . Zhang, P. Xiao, X. Sun, and F. Wu, “Better and faster: Adap- tive event conversion for event-based object detection,” in Proceedings of the AAAI Conference on Artificial Intelligence , vol. 37, no. 2, 2023, pp. 2056–2064

    2023

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.