Pith. sign in

REVIEW 6 major objections 4 minor 62 references

Unsupervised Sparse Coding-based Spiking Neural Network for Real-time Spike Sorting

T0 review · 6 major / 4 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read A two-layer spiking sparse coder can sort tetrode spikes online, unsupervised, and on-chip, with 2-bit graded spikes beating LIF neurons on drifting data.

desk verdict A genuine Loihi 2 spike-sorting implementation with graded spikes, undermined by an abstract whose numbers don't match the paper's own tables. read the letter →

arxiv 2506.24041 v1 pith:2I2TF3XN submitted 2025-06-30 cs.NE cs.LG

Alexis Melot , Sean U.N. Wood , Yannick Coffinier , Pierre Yger , Fabien Alibart This is my paper · ORCID
classification cs.NEcs.LG
keywords spikingneuralnetworkspikesortingsparsecodinglocallycompetitivealgorithmunsupervisedlearningneuromorphiccomputinggradedspikesbrain-machineinterfaces
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

This paper introduces the Neuromorphic Sparse Sorter (NSS), a two-layer spiking neural network that sorts extracellular tetrode spikes without labels and without offline batch training. The central claim is that a spiking version of the Locally Competitive Algorithm, running on a neuromorphic chip that supports graded spikes, can give the accuracy of conventional sparse-coding spike sorters while staying fast and low-power enough for edge use in brain-machine interfaces. With 2-bit graded spikes, NSS on Loihi 2 outperforms the same network built from leaky integrate-and-fire neurons, reaching an F1-score of 77% on a drifting real recording at 8.6 mW dynamic power and 0.25 ms per waveform, though the timing figure excludes off-chip input/output latency. If the claim holds, NSS is a path to on-chip, in situ spike sorting that transmits only sorted action potentials rather than raw multichannel data, and the paper is explicit that its accuracy still trails offline sorters such as Kilosort on several benchmarks.

What carries the argument

The load-bearing object is the two-layer spiking Locally Competitive Algorithm network. LCA is a recurrent neural circuit that solves the LASSO sparse-coding problem by letting neurons compete through lateral inhibition; the first NSS layer encodes each 120-dimensional spike waveform as a sparse code, and the second layer clusters by taking the index of the most active neuron. A custom neuron model applies the Temporally Diffused Quantizer (TDQ) to the rectified softshrink activation, quantizing continuous coefficients into N-bit graded spikes while carrying quantization error forward in time. The bit-width S is the single knob that trades temporal sparsity and energy against sorting accuracy, and S=2 is chosen as the sweet spot.

What would settle it

Run the same NSS-2bit configuration on Loihi 2 and measure wall-clock latency from spike waveform arrival to sorted label, including the documented 117.5 ms I/O transfer; if that latency exceeds the roughly 100-170 ms interval between spikes of a 6-10 Hz neuron, the system is not real-time for the tested recordings.

Watch

Extended reading notes

Core claim

The paper's claim, on its own terms, is that a compact two-layer LCA network can replace the feature-extraction and clustering stages of a spike-sorting pipeline, and that quantized graded spikes make that network implementable on digital neuromorphic hardware. The first layer learns a dictionary of waveform atoms from incoming spike waveforms and outputs sparse codes; the second layer acts as a clustering stage whose label is the index of its most active neuron. A custom neuron model applies the Temporally Diffused Quantizer to the rectified softshrink activation, converting continuous coefficients into graded spikes while propagating quantization error forward in time, and the authors show that 2-bit spikes capture most of the accuracy gain of higher precision while keeping temporal sparsity high. On Loihi 2 the network occupies two neurocores and runs in the single-digit milliwatt dynamic-power range, with the largest F1 gain over LIF neurons appearing on the recording with biological drift. The learning rule is Hebbian-like and the design is meant to run online and unsupervised, though the Loihi 2 runs use weights trained off-chip and frozen because the chip does not support the layerwise learning rule.

Load-bearing premise

The real-time claim stands on counting only neurocore computation time and excluding the paper's own measured 117.5 ms input/output transfer per spike waveform, which would otherwise push each inference to about 118 ms.

Editorial extensions

If this is right

  • On the drifting real recording, the 2-bit graded-spike NSS on Loihi 2 raised F1 after drift from 59.0% for the LIF version to 71.4%, with dynamic power rising only from 7.95 mW to 8.60 mW and neurocore inference time at 0.26 ms per waveform.
  • NSS is competitive with WaveClus3 and above PCA+KMeans across most synthetic and real tetrode recordings, despite processing online and without labels.
  • Because the second layer labels by the most active neuron and needs no cluster count, NSS avoids the parameterization requirement of KMeans-based pipelines.
  • The whole network uses two neurocores of a Loihi 2 chip and about 30,100 synapses, and the authors estimate that a hypothetical 64-channel version would use about 7.6 million synapses, still within one chip.
  • On several benchmark datasets, NSS remains less accurate than offline sorters such as Kilosort and Spyking-Circus, a trade-off the authors attribute to the simpler pipeline.

Reading between the lines

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

  • If the off-chip I/O latency is fixed in future hardware, the same architecture would run end-to-end well inside a typical inter-spike interval, since the neurocore computation itself is under 0.4 ms; the prototype's 117.5 ms transfer is not intrinsic to the algorithm.
  • The bit-width dial generalizes: any LCA-based neuromorphic task could use the same TDQ quantization to sweep from continuous accuracy to binary spiking efficiency by changing one parameter.
  • A testable extension is to keep the dictionary learning running during the drift phase rather than freezing weights; the paper's online-adaptation rationale predicts slower F1 decay than the frozen Loihi implementation.
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

6 major / 4 minor

Summary. The manuscript proposes the Neuromorphic Sparse Sorter (NSS), a two-layer spiking neural network based on the Locally Competitive Algorithm (LCA) for unsupervised spike sorting of tetrode recordings. NSS uses a first LCA layer for sparse feature extraction and a second LCA layer for clustering, with a rectified softshrink activation and the Temporally Diffused Quantizer (TDQ) to produce graded spikes. The method is evaluated on five synthetic and four real tetrode recordings against PCA+KMeans, WaveClus3, and other sorters, and a Loihi 2 implementation with a custom neuron model is benchmarked against a LIF version. The central claim is that NSS is a real-time, low-power, unsupervised sorter whose 2-bit graded-spike version improves spike-sorting F1 from 59.0% to 71.4% on a drifting recording at small additional power.

Significance. If the claims were fully supported, the paper would be a valuable demonstration of a digital neuromorphic implementation of an unsupervised sparse-coding spike sorter, and the graded-spike trade-off would be practically useful for edge BMI systems. The authors release their code, provide CPU and hardware measurements, compare against several established baselines, and include sensitivity analyses, which are strengths of the submission. However, as submitted, the headline contributions are weakened by inconsistencies between the abstract and the body, an excluded I/O latency that undermines the real-time claim, unclear power units, and a hyperparameter selected on the evaluation data.

major comments (6)
  1. [Abstract and Table 3] The abstract reports an F1-score of 77% with a +10% improvement, +1.65 mW, and 0.25 ms (+60 µs) per inference, but Table 3 reports 71.4% F1 after drift and 8.60 versus 7.95 (units unclear) dynamic power for the same comparison; the 77% value appears nowhere in the body. Please correct the abstract to match the measured values and state the exact evaluation conditions (before/after drift, number of waveforms, and confidence intervals).
  2. [Sections 2.3.4 and 3.3] The real-time claim is not demonstrated for the system as built. Section 2.3.4 states that the measured I/O transfer time for one spike waveform and its outputs is 117.5 ms, and that all inference time measurements exclude this latency; Section 3.3 repeats that the sub-0.4 ms processing time excludes I/O communication. For the reported 6–10 Hz firing rates, the inter-spike interval is roughly 100–166 ms, so adding the 117.5 ms I/O time to the 0.25 ms neurocore time yields approximately 118 ms per waveform, which is at or above the biological event interval. The host-side detection and alignment stages are also excluded. Please provide an end-to-end latency budget that includes detection, alignment, I/O, and inference, or remove the real-time claim from the abstract and conclusion.
  3. [Table 3 and Section 3.3] The power and energy figures are internally inconsistent. Table 3 labels dynamic power in µW and reports values of 5.20–18.30, while the abstract states 8.6 mW (+1.65 mW) and Section 3.3 refers to “mW/channel”. If the measured values are µW, the abstract overstates power by a factor of 1000; if the values are mW, the table header is wrong. The dynamic energy and energy-delay-product columns also mix units in a way that makes the reported EDP values difficult to reproduce. Please state the exact units for every quantity and compute energy and EDP from a single consistent set of base units.
  4. [Section 3.1] The choice of S=2 is made after observing the F1-score improvements on the evaluation datasets, including the real drifting recording TR1, in Figure 4, and the same S=2 is then used for the headline Loihi 2 comparison in Table 3. Since the spike bit-width is a hyperparameter selected on test data, the reported 71.4% versus 59.0% advantage over NSS-LIF may be optimistic. Please select S on the held-out TS0 dataset or report the full bit-width sweep on the Loihi benchmark with estimates of variability.
  5. [Sections 2.3.4 and 3.3] The statements that NSS “learns to sort” and “operates entirely unsupervised” apply to the CPU-based algorithm, but the Loihi 2 demonstration does not include on-chip learning: Section 2.3.4 states that NSS was trained offline on a CPU with S=8 and the frozen weights were then transferred to the chip. Moreover, the custom neuron model was programmed manually, and no on-chip dictionary updates were performed. Please distinguish the offline-trained hardware inference from the online learning algorithm and state this limitation explicitly in the abstract and conclusion.
  6. [Sections 3.2 and 4, Table 4] The broad claim that NSS outperforms established pipelines such as WaveClus3 and PCA+KMeans is not supported by the reported results. Section 3.2 states that WaveClus3 is better on the drifting real recording, and Table 4 shows that NSS-2bit is below Tridesclous, Spyking-Circus, and Kilosort on several datasets. Please restrict the performance claims to the specific regimes where they are supported, or qualify them as “competitive” rather than “superior”.
minor comments (4)
  1. [Table 1] The real-world dataset rows list “TR2” twice, with different spike rates and durations; this appears to be a labeling error, and the mapping between dataset names and the TR1–TR4 labels used in the text should be clarified.
  2. [Introduction and References] Reference [40] (Gold et al., 2006) is cited for the statement that Loihi 2 and SpiNNaker support multi-bit spikes, but that reference is a modeling study of extracellular action potentials; the citation appears to be incorrect and should be replaced with the appropriate hardware references.
  3. [Table 3] The after-drift F1-score for TR1 is computed on only the last 100 spike waveforms; given the small sample, confidence intervals or a bootstrap estimate should be reported for the headline F1 comparison.
  4. [Section 2.2.5] Equation (4) and the surrounding text contain typesetting artifacts (missing characters and undefined notation such as “∙” and “v(t)”) that make the TDQ recurrence difficult to read; the equation should be rewritten cleanly with all variables defined.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: NSS is benchmarked against external baselines, and its LCA/TDQ components rest on independent prior work; the flagged I/O-latency exclusion is a validity limitation, not a circular reduction.

full rationale

The derivation of NSS does not define its target results in terms of its inputs. The network uses standard LCA dynamics (Eq. 2) and dictionary update (Eq. 3) from Rozell et al. and the TDQ quantizer from Voelker et al., which are independent prior results rather than a self-citation chain. The authors' own prior LCA study is cited twice ([34]) to motivate LCA feature extraction and a one-time-overcomplete dictionary, but those citations are not the load-bearing source of the headline claim, which is supported by comparisons against WaveClus3, PCA+KMeans, Kilosort, Tridesclous, and Spyking-Circus on external datasets. The main flagged limitation is in Section 2.3.4: 'All subsequent inference time measurements reflect only the neurocores computation time and exclude I/O latency,' with an average I/O transfer of 117.5 ms per waveform. This undermines the abstract's real-time framing, but it is an evaluation-boundary choice, not a circular derivation. The selection of S=2 in Section 3.1 from Figure 4 ('Based on these findings, S=2 was selected for the remainder of the study') is model selection on the same datasets, which weakens external validity, but the reported F1 scores are measured outputs after that choice, not quantities forced by construction. No equation or definition reduces a claimed prediction to a fitted parameter or to a self-citation.

Assumptions & free parameters 10 free parameters · 6 assumptions · 0 invented entities

The central claim does not rest on new physical entities; no invented entities are introduced. It does depend on a set of tuned hyperparameters, the most consequential being the post-hoc choice of spike bit-width S=2, and on prior results for LCA convergence and TDQ differentiation. The learning rule and two-layer online schedule are validated empirically rather than by proof in this paper.

free parameters (10)
  • lambda (LCA sparsity threshold) = 0.03
    Optimized on TS0 with Optuna (Section 2.3.1, Table 2); controls sparsity and denoising behavior of both LCA layers.
  • tau (leak time constant) = 2 ms
    Optimized hyperparameter (Table 2); sets the integration time of LCA neurons.
  • eta (learning rate schedule) = 0.07 -> 0.01
    Scheduled learning rate for dictionary updates (Table 2); strong learning for first 60 s, slower thereafter.
  • delta_t (discrete time step) = 0.1 ms
    Time step used for LCA iteration and spike timing (Table 2).
  • Number of time steps per SW = 200 -> 50
    Scheduled iteration count used during strong learning and after convergence (Table 2).
  • M1 (LCA1 dictionary size) = 120
    Chosen equal to input dimension based on the authors' prior finding that 10x overcomplete adds less than 0.1% F1 (Section 2.3.1).
  • M2 (LCA2 dictionary size) = 10
    Output layer size larger than the number of bioneurons to demonstrate no prior knowledge of cluster count (Section 2.3.1).
  • S (spike bit-width) = 2
    Selected after sweeping S on simulated and real test recordings; drives the headline F1 and power trade-off (Section 3.1).
  • lambda_LIF (LIF threshold) = 1.06
    Optimized experimentally on TS0 for the LIF comparison baseline (Section 2.3.4).
  • Learning-rule noise variance = 0.03
    Variance of the zero-mean Gaussian noise added to the dictionary update for stability (Section 2.2.3).
assumptions (6)
  • standard math LCA converges to the LASSO solution of Eq. (1) for the continuous activation function.
    Relied on in Section 2.2.2 (citing Rozell 2008 and Tang 2017); the paper does not re-prove convergence for its two-layer or quantized setting.
  • domain assumption The dictionary update rule of Eq. (3) is effective gradient descent on the sparse-coding objective and remains stable with the scheduled online learning.
    Used in Section 2.2.3; stability and convergence are demonstrated empirically, not analytically.
  • standard math The TDQ quantizer has derivative equal to 1, so training with quantized activations remains valid.
    Inherited in Section 2.2.5 from Voelker et al.; assumed for the graded-spike neuron model on Loihi 2.
  • domain assumption Synthetic templates generated by the spikeinterface simulator default model are representative of real tetrode signals.
    Defines the benchmark in Section 2.1; results on synthetic data are used for hyperparameter selection and main comparisons.
  • domain assumption Best-unit matching of ground-truth and inferred spike trains with a 1 ms tolerance yields a valid F1 measure of sorting quality.
    Evaluation protocol in Section 2.3.3; with only one ground-truth neuron in real recordings, F1 covers only that neuron.
  • domain assumption Fixed-point conversion of trained weights to Loihi 2 preserves sorting behavior at the reported F1 levels.
    Implementation step in Section 2.3.4; no numerical error analysis is provided.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Unsupervised Sparse Coding-based Spiking Neural Network for Real-time Spike Sorting." pith.science (2026). https://pith.science/paper/2I2TF3XN

@misc{pith2026250624041,
  author       = {Pith},
  title        = {Pith review of: Unsupervised Sparse Coding-based Spiking Neural Network for Real-time Spike Sorting},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/2I2TF3XN}},
  note         = {Machine review of arXiv:2506.24041}
}
read the original abstract

Spike sorting is a crucial step in decoding multichannel extracellular neural signals, enabling the identification of individual neuronal activity. A key challenge in brain-machine interfaces (BMIs) is achieving real-time, low-power spike sorting at the edge while keeping high neural decoding performance. This study introduces the Neuromorphic Sparse Sorter (NSS), a compact two-layer spiking neural network optimized for efficient spike sorting. NSS leverages the Locally Competitive Algorithm (LCA) for sparse coding to extract relevant features from noisy events with reduced computational demands. NSS learns to sort detected spike waveforms in an online fashion and operates entirely unsupervised. To exploit multi-bit spike coding capabilities of neuromorphic platforms like Intel's Loihi 2, a custom neuron model was implemented, enabling flexible power-performance trade-offs via adjustable spike bit-widths. Evaluations on simulated and real-world tetrode signals with biological drift showed NSS outperformed established pipelines such as WaveClus3 and PCA+KMeans. With 2-bit graded spikes, NSS on Loihi 2 outperformed NSS implemented with leaky integrate-and-fire neuron and achieved an F1-score of 77% (+10% improvement) while consuming 8.6mW (+1.65mW) when tested on a drifting recording, with a computational processing time of 0.25ms (+60 us) per inference.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

62 extracted references · 40 canonical work pages

  1. [1]

    Concept cells: the building blocks of declarative memory functions,

    R. Q. Quiroga, “Concept cells: the building blocks of declarative memory functions,” Nature Reviews Neuroscience, 2012 13:8, vol. 13, no. 8, pp. 587–597, 2012, doi: 10.1038/nrn3251

    doi:10.1038/nrn3251 2012

  2. [2]

    A Neuromorphic Prosthesis to Restore Communication in Neuronal Networks,

    S. Buccelli et al., “A Neuromorphic Prosthesis to Restore Communication in Neuronal Networks,” iScience, vol. 19, pp. 402–414, 2019, doi: 10.1016/j.isci.2019.07.046

    work page doi:10.1016/j.isci.2019.07.046 2019

  3. [3]

    Unsupervised spike detection and sorting with wavelets and superparamagnetic clustering,

    R. Q. Quiroga, Z. Nadasdy, and Y. Ben-Shaul, “Unsupervised spike detection and sorting with wavelets and superparamagnetic clustering,” Neural Computation, vol. 16, no. 8, pp. 1661–1687, 2004, doi: 10.1162/089976604774201631

    work page doi:10.1162/089976604774201631 2004

  4. [4]

    Fully integrated silicon probes for high- density recording of neural activity,

    J. J. Jun et al., “Fully integrated silicon probes for high- density recording of neural activity,” Nature, vol. 551, no. 7679, pp. 232–236, 2017, doi: 10.1038/NATURE24636

    doi:10.1038/nature24636 2017

  5. [5]

    Neuropixels 2.0: A miniaturized high-density probe for stable, long-term brain recordings

    N. A. Steinmetz et al., “ Neuropixels 2.0: A miniaturized high-density probe for stable, long-term brain recordings ”, Science, vol. 372, no 6539, 2021, doi: 10.1126/SCIENCE.ABF4588

    doi:10.1126/science.abf4588 2021

  6. [6]

    Kilosort: realtime spike-sorting for extracellular electrophysiology with hundreds of channels

    M. Pachitariu, N. Steinmetz, S. Kadir, M. Carandini, and H. K. D., “ Kilosort: realtime spike-sorting for extracellular electrophysiology with hundreds of channels ”, bioRxiv, p. 61481, 2016, doi: 10.1101/061481

    work page doi:10.1101/061481 2016

  7. [7]

    A spike sorting toolbox for up to thousands of electrodes validated with ground truth recordings in vitro and in vivo

    P. Yger et al., “ A spike sorting toolbox for up to thousands of electrodes validated with ground truth recordings in vitro and in vivo ”, Elife, vol. 7, 2018, doi: 10.7554/ELIFE.34518

    work page doi:10.7554/elife.34518 2018

  8. [8]

    Spikeinterface, a unified framework for spike sorting

    A. P. Buccino et al., “ Spikeinterface, a unified framework for spike sorting ”, Elife, vol. 9, p. 1-24, 2020, doi: 10.7554/eLife.61834

    work page doi:10.7554/elife.61834 2020

Show all 62 references
  1. [9]

    A Fully Automated Approach to Spike Sorting

    J. E. Chung et al., “ A Fully Automated Approach to Spike Sorting ”, Neuron, vol. 95, no 6, p. 1381-1394.e6, september 2017, doi: 10.1016/J.NEURON.2017.08.030

    2017 doi

  2. [10]

    A fully automatic multichannel neural spike sorting algorithm with spike reduction and positional feature

    Z. Mohammadi, D. J. Denman, A. Klug, and T. C. Lei, “ A fully automatic multichannel neural spike sorting algorithm with spike reduction and positional feature ”, J Neural Engineering, vol. 21, no 4, august 2024, doi: 10.1088/1741-2552/ad647d

    2024 doi

  3. [11]

    Thermal impact of an active 3-D microelectrode array implanted in the brain

    S. Kim, P. Tathireddy, R. A. Normann, and F. Solzbacher, “ Thermal impact of an active 3-D microelectrode array implanted in the brain ”, IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. 15, no 4, p. 493-501, december 2007, doi: 10.1109/TNSRE.2007.908429

    2007

  4. [12]

    An area-efficient 128-channel spike sorting processor for real-time neural recording with 0.175 μ W/Channel in 65-nm CMOS

    A. T. Do, S. M. A. Zeinolabedin, D. Jeon, D. Sylvester, and T. T. H. Kim, “ An area-efficient 128-channel spike sorting processor for real-time neural recording with 0.175 μ W/Channel in 65-nm CMOS ”, IEEE Transaction of Very Large Scale Integration System, vol. 27, no 1, p. 1...

    2019

  5. [13]

    A New Spike Sorting Algorithm Based on Continuous Wavelet Transform and Investigating Its Effect on Improving Neural Decoding Accuracy

    A. Soleymankhani and V. Shalchyan, “ A New Spike Sorting Algorithm Based on Continuous Wavelet Transform and Investigating Its Effect on Improving Neural Decoding Accuracy ”, Neuroscience, vol. 468, p. 139-148, 2021, doi: 10.1016/J.NEUROSCIENCE.2021.05.036

    2021 doi

  6. [14]

    A new approach to spike sorting for multi-neuronal activities recorded with a tetrode - how ICA can be practical

    S. Takahashi, Y. Anzai, and Y. Sakurai, “ A new approach to spike sorting for multi-neuronal activities recorded with a tetrode - how ICA can be practical ”, Neuroscience Research, vol. 46, no 3, p. 265-272, july 2003, doi: 10.1016/S0168-0102(03)00103-2

    2003 doi

  7. [15]

    Veerabhadrappa, M

    R. Veerabhadrappa, M. Ul Hassan, J. Zhang, and A. Bhatti, “ Compatibility Evaluation of Clustering Algorithms for Contemporary Extracellular Neural Spike Sorting Frontiers in Neuroscience, vol. 14, june 2020, doi: 10.3389/fnsys.2020.00034

    2020

  8. [16]

    An automatic spike sorting algorithm based on adaptive spike detection and a mixture of skew-t distributions

    R. Toosi, M. A. Akhaee, and M. R. A. Dehaqani, “ An automatic spike sorting algorithm based on adaptive spike detection and a mixture of skew-t distributions ”, Scientific Reports, 2021 11:1, vol. 11, no 1, p. 1-18, 2021, doi: 10.1038/s41598-021-93088-w

    2021 doi

  9. [17]

    Spike sorting: new trends and challenges of the era of high-density probes

    A. P. Buccino, S. Garcia, and P. Yger, “ Spike sorting: new trends and challenges of the era of high-density probes ”, Progress in Biomedical Engineering, vol. 4, no 2, april 2022, doi: 10.1088/2516-1091/ac6b96

    2022 doi

  10. [18]

    A Multi-Channel Spike Sorting Processor with Accurate Clustering Algorithm Using Convolutional Autoencoder

    C. Seong, W. Lee, and D. Jeon, “ A Multi-Channel Spike Sorting Processor with Accurate Clustering Algorithm Using Convolutional Autoencoder ”, IEEE Transactions on Biomedical Circuits and System, vol. 15, no 6, p. 1441-1453, december 2021, doi: 10.1109/TBCAS.2021.3134660

    2021

  11. [19]

    Removing Noise from Extracellular Neural Recordings Using Fully Convolutional Denoising Autoencoders

    C. Kechris, A. Delitzas, V. Matsoukas, and P. C. Petrantonakis, “ Removing Noise from Extracellular Neural Recordings Using Fully Convolutional Denoising Autoencoders ”, ArXiv, september 2021, doi: 10.1109/EMBC46164.2021.9630585. [21]J. Rokai, I. Ulbert, and G. Márton, “ Edge ...

    2021

  12. [22]

    DualSort: online spike sorting with a running neural network

    L. M. Meyer, F. Samann, and T. Schanze, “ DualSort: online spike sorting with a running neural network ”, Journal of Neuralal Engineering, vol. 20, no 5, p. 056031, october 2023, doi: 10.1088/1741- 2552/ACFB3A

    2023 doi

  13. [23]

    Memristor networks for real-time neural activity analysis

    X. Zhu, Q. Wang, and W. D. Lu, “ Memristor networks for real-time neural activity analysis ”, Nature Communication, vol. 11, no 1, 2020, doi: 10.1038/s41467-020-16261-1

    2020 doi

  14. [24]

    Real-time encoding and compression of neuronal spikes by metal-oxide memristors

    I. Gupta, A. Serb, A. Khiat, R. Zeitler, S. Vassanelli, and T. Prodromakis, “ Real-time encoding and compression of neuronal spikes by metal-oxide memristors ”, Nature Communication, vol. 7, 2016, doi: 10.1038/ncomms12805

    2016 doi

  15. [25]

    A critical survey of STDP in Spiking Neural Networks for Pattern Recognition

    A. Vigneron and J. Martinet, “A critical survey of STDP in Spiking Neural Networks for Pattern Recognition”, International Joint Conference on Neural Networks, july 2020, doi: 10.1109/IJCNN48605.2020.9207239

    2020

  16. [26]

    The nonlinear PCA learning rule in independent component analysis

    E. Oja, “ The nonlinear PCA learning rule in independent component analysis ”, Neurocomputing, vol. 17, no 1, p. 25-45, september 1997, doi: 10.1016/S0925- 2312(97)00045-3

    1997 doi

  17. [27]

    Spiking neural networks based on OxRAM synapses for real-time unsupervised spike sorting

    T. Werner et al., “ Spiking neural networks based on OxRAM synapses for real-time unsupervised spike sorting ”, Frontiers in Neuroscience, vol. 10, november 2016, doi: 10.3389/fnins.2016.00474

    2016

  18. [28]

    An Attention-Based Spiking Neural Network for Unsupervised Spike-Sorting

    M. Bernert and B. Yvert, “ An Attention-Based Spiking Neural Network for Unsupervised Spike-Sorting ”, International Journal of Neural Systems, vol. 29, no 8, october 2019, doi: 10.1142/S0129065718500594

    2019 doi

  19. [29]

    Advancing Neuromorphic Computing with Loihi: A Survey of Results and Outlook

    M. Davies et al., “ Advancing Neuromorphic Computing with Loihi: A Survey of Results and Outlook ”, Proceedings of the IEEE, vol. 109, no 5, p. 911-934, 2021, doi: 10.1109/JPROC.2021.3067593

    2021

  20. [30]

    Loihi: A Neuromorphic Manycore Processor with On-Chip Learning

    M. Davies et al., “ Loihi: A Neuromorphic Manycore Processor with On-Chip Learning ”, IEEE Micro, vol. 38, no 1, p. 82-99, 2018, doi: 10.1109/MM.2018.112130359

    2018

  21. [31]

    Designing a bidirectional, adaptive neural interface incorporating machine learning capabilities and memristor-enhanced hardware

    S. Shchanikov et al., “ Designing a bidirectional, adaptive neural interface incorporating machine learning capabilities and memristor-enhanced hardware ”, Chaos Solitons Fractals, vol. 142, 2021, doi: 10.1016/j.chaos.2020.110504

    2021

  22. [32]

    Efficient Neuromorphic Signal Processing with Loihi 2

    G. Orchard et al., “ Efficient Neuromorphic Signal Processing with Loihi 2 ”, IEEE Workshop on Signal Processing Systems, SiPS: Design and Implementation,, 2021, p. 254-259. doi: 10.1109/SiPS52927.2021.00053

    2021

  23. [33]

    SpiNNaker: A 1-W 18-core system- on-chip for massively-parallel neural network simulation

    E. Painkras et al., “ SpiNNaker: A 1-W 18-core system- on-chip for massively-parallel neural network simulation ”, IEEE Journal of Solid-State Circuits, vol. 48, no 8, p. 1943-1953, 2013, doi: 10.1109/JSSC.2013.2259038

    1943

  24. [34]

    Sparse Coding-based Multichannel Spike Sorting with the Locally Competitive Algorithm

    A. Mélot, F. Alibart, P. Yger, and S. U. N. Wood, “ Sparse Coding-based Multichannel Spike Sorting with the Locally Competitive Algorithm ”, dans 2023 IEEE Biomedical Circuits and Systems Conference (BioCAS), Toronto, oct. 2023, doi : 10.1109/BioCAS58349.2023.10388594

    2023

  25. [35]

    Sparse coding via thresholding and local competition in neural circuits

    C. J. Rozell, D. H. Johnson, R. G. Baraniuk, and B. A. Olshausen, “ Sparse coding via thresholding and local competition in neural circuits ”, Neural Computation, vol. 20, no 10, p. 2526-2563, 2008, doi: 10.1162/NECO.2008.03-07-486

    2008 doi

  26. [36]

    Unsupervised Spike Sorting for Large-Scale, High-Density Multielectrode Arrays

    G. Hilgen et al., “ Unsupervised Spike Sorting for Large-Scale, High-Density Multielectrode Arrays ”, Cell Reports, vol. 18, no 10, p. 2521-2532, 2017, doi: 10.1016/J.CELREP.2017.02.038

    2017 doi

  27. [37]

    Spike sorting for large, dense electrode arrays

    C. Rossant et al., “ Spike sorting for large, dense electrode arrays ”, Nature Neuroscience, 2016 19:4, vol. 19, no 4, p. 634-641, 2016, doi: 10.1038/nn.4268

    2016 doi

  28. [38]

    Sparse Coding on Stereo Video for Object Detection

    S. Y. Lundquist, M. Mitchell, and G. T. Kenyon, “ Sparse Coding on Stereo Video for Object Detection ” arXiv, 2017, doi: 10.48550/arxiv.1705.07144

    2017 doi arXiv

  29. [39]

    Feature Extraction Based on Sparse Coding Approach for Hand Grasp Type Classification

    J. Samkunta, P. Ketthong, N. T. Mai, M. A. S. Kamal, I. Murakami, and K. Yamada, “ Feature Extraction Based on Sparse Coding Approach for Hand Grasp Type Classification ”, Algorithms, vol. 17, no 6, juin 2024, doi: 10.3390/a17060240

    2024 doi

  30. [40]

    On the origin of the extracellular action potential waveform: A modeling study

    C. Gold, D. A. Henze, C. Koch, and G. Buzsáki, “ On the origin of the extracellular action potential waveform: A modeling study ”, Journal of Neurophysioly, vol. 95, no 5, p. 3113-3128, mai 2006, doi: 10.1152/jn.00979.2005

    2006

  31. [41]

    De-Noising by Soft-Thresholding

    D. L. Donoho, “ De-Noising by Soft-Thresholding ”, IEEE Transactions on Information Theory, vol. 41, no 3, p. 613, august 1995, doi: 10.1109/18.382009

    1995 doi

  32. [42]

    M. H. Malik, M. Saeed, et A. M. Kamboh, « Automatic threshold optimization in nonlinear energy operator based spike detection », Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS, vol. 2016-October, p. 774 777, oct...

    2016

  33. [43]

    A. T. Do et K. S. Yeo, « A hybrid NEO-based spike detection algorithm for implantable brain-IC interface applications », Proceedings - IEEE International 15 Symposium on Circuits and Systems, p. 2393 2396, 2014, doi: 10.1109/ISCAS.2014.6865654

    2014

  34. [44]

    NeuSort: an automatic adaptive spike sorting approach with neuromorphic models

    Y., Hang, Y. Qi, and G. Pan. "NeuSort: an automatic adaptive spike sorting approach with neuromorphic models." Journal of Neural Engineering 20, no. 5, 2023, doi: 10.1088/1741-2552/acf61d

    2023 doi

  35. [45]

    Comparison of low-power biopotential processors for on-the-fly spike detection

    G. Gagnon-Turcotte, C. O. D. Camaro, and B. Gosselin, “ Comparison of low-power biopotential processors for on-the-fly spike detection ”, Proceedings - IEEE International Symposium on Circuits and Systems, vol. 2015- July, p. 802-805, july 2015, doi: 10.1109/ISCAS.2015.7168755

    2015

  36. [46]

    Sparse coding with an overcomplete basis set: A strategy employed by V1?

    B. A. Olshausen and D. J. Field, “ Sparse coding with an overcomplete basis set: A strategy employed by V1? ”, Vision Research, vol. 37, no 23, p. 3311-3325, 1997, doi: 10.1016/S0042-6989(97)00169-7

    1997 doi

  37. [47]

    Sparse Coding by Spiking Neural Networks: Convergence Theory and Computational Results

    P. T. P. Tang, T.-H. Lin, and M. Davies, “ Sparse Coding by Spiking Neural Networks: Convergence Theory and Computational Results ”, ArXiv, 2017, doi: 10.48550/arXiv.1705.05475

    2017 doi arXiv

  38. [48]

    K-SVD: An algorithm for designing overcomplete dictionaries for sparse representation

    M. Aharon, M. Elad, and A. Bruckstein, “ K-SVD: An algorithm for designing overcomplete dictionaries for sparse representation ”, IEEE Transactions on Signal Processing, vol. 54, no 11, p. 4311-4322, 2006, doi: 10.1109/TSP.2006.881199

    2006

  39. [49]

    Efficient sparse coding algorithms

    H. Lee, A. B. Rajat, R. Andrew, and Y. Ng, “ Efficient sparse coding algorithms ”, Advances in Neural Inference Process System, vol. 19, 2006, ISBN: 9780262256919

    2006

  40. [50]

    Learning Fast Approximations of Sparse Coding

    K. Gregor and Y. Lecun, “ Learning Fast Approximations of Sparse Coding ”, Proceedings of the 27th International Conference on Machine Learning, 2010, doi: 10.5555/3104322.3104374

    2010

  41. [51]

    Spiking Sparse Coding Algorithm with Reduced Inhibitory Feedback Weights,

    M. M. Hasan and J. Holleman, “Spiking Sparse Coding Algorithm with Reduced Inhibitory Feedback Weights,” IEEE 63rd International Midwest Symposium on Circuits and Systems (MWSCAS), Springfield, MA, USA, 2020, pp. 1040–1043, 2020, doi: 10.1016/j.neucom.2017.05.016

    2020 doi

  42. [52]

    Sparse coding using the locally competitive algorithm on the truenorth neurosynaptic system

    K. L. Fair, D. R. Mendat, A. G. Andreou, C. J. Rozell, J. Romberg, and D. V. Anderson, “ Sparse coding using the locally competitive algorithm on the truenorth neurosynaptic system ”, Frontiers in Neuroscience, vol. 13, july 2019, doi: 10.3389/fnins.2019.00754

    2019

  43. [53]

    Highly overcomplete sparse coding

    B. A. Olshausen, “ Highly overcomplete sparse coding ”, Human Vision and Electronic Imaging, march 2013, p. 86510S. doi: 10.1117/12.2013504

    2013 doi

  44. [54]

    Neural- based approach to perceptual sparse coding of audio signals

    R. Pichevar, H. Najaf-Zadeh, and F. Mustiere, “ Neural- based approach to perceptual sparse coding of audio signals ”, Proceedings of the International Joint Conference on Neural Networks, 2010, doi: 10.1109/IJCNN.2010.5596912

    2010

  45. [55]

    A hybrid wavelet convolution network with sparse-coding for image super-resolution,

    X. Gao and H. Xiong, "A hybrid wavelet convolution network with sparse-coding for image super-resolution," 2016 IEEE International Conference on Image Processing, Phoenix, AZ, USA, 2016, pp. 1439-1443, doi: 10.1109/ICIP.2016.7532596. [56]T. Xiong et al., “ An unsupervised comp...

    2016

  46. [57]

    Adaptive Approach For Sparse Representations Using The Locally Competitive Algorithm For Audio

    S. Bahadi, J. Rouat, and É. Plourde, “ Adaptive Approach For Sparse Representations Using The Locally Competitive Algorithm For Audio ”, IEEE International Workshop on Machine Learning for Signal Processing, 2021, doi: 10.1109/MLSP52302.2021.9596348

    2021

  47. [58]

    The Sparse Manifold Transform

    Y. Chen, D. M. Paiton, and B. A. Olshausen, “ The Sparse Manifold Transform ”, ArXiv, 2018, doi: 10.48550/arXiv.1806.08887

    2018 doi arXiv

  48. [59]

    Fast and Accurate Sparse Coding of Visual Stimuli with a Simple, Ultralow-Energy Spiking Architecture

    W. Woods and C. Teuscher, “ Fast and Accurate Sparse Coding of Visual Stimuli with a Simple, Ultralow-Energy Spiking Architecture ”, IEEE Transaction on Neural Network and Learning Systems, vol. 30, no 7, p. 2173-2187, july 2019, doi: 10.1109/TNNLS.2018.2878002

    2019

  49. [60]

    Dictionary Learning with Accumulator Neurons

    G. Parpart Yijing Watkins Carlos González et al., “ Dictionary Learning with Accumulator Neurons”, ArXiv, 2022, doi: 10.48550/arXiv.2205.15386

    2022 doi arXiv

  50. [61]

    A Spike in Performance: Training Hybrid-Spiking Neural Networks with Quantized Activation Functions

    A. R. Voelker, D. Rasmussen, and C. Eliasmith, “ A Spike in Performance: Training Hybrid-Spiking Neural Networks with Quantized Activation Functions ”, ArXiv, february 2020, doi: 10.48550/arXiv.2002.03553

    2020 doi arXiv

  51. [62]

    Performance evaluation of PCA-based spike sorting algorithms

    D. A. Adamos, E. K. Kosmidis, and G. Theophilidis, “ Performance evaluation of PCA-based spike sorting algorithms ”, Comput Methods Programs Biomed, vol. 91, no 3, p. 232-244, september 2008, doi: 10.1016/j.cmpb.2008.04.011

    2008 doi

  52. [63]

    Innovative- Methodology: A novel and fully automatic spike-sorting implementation with variable number of features

    F. J. Chaure, H. G. Rey, and R. Q. Quiroga, “ Innovative- Methodology: A novel and fully automatic spike-sorting implementation with variable number of features ”, Journal of Neurophysioly, vol. 120, p. 1859-1871, 2018, doi: 10.1152/jn.00339.2018

    2018

  53. [64]

    Watkins, A

    Y. Watkins, A. Thresher, P. F. Schultz, A. Wild, A. Sornborger, and G. T. Kenyon, « Unsupervised 16 dictionary learning via a spiking locally competitive algorithm », ACM International Conference Proceeding Series, Association for Computing Machinery, 2019. doi: 10.1145/335426...

    2019

  54. [65]

    Convolutional Dictionary Learning with Grid Refinement

    A. H. Song, F. J. Flores, and D. Ba, “ Convolutional Dictionary Learning with Grid Refinement ”, IEEE Transactions on Signal Processing, vol. 68, p. 2558-2573, 2020, doi: 10.1109/TSP.2020.2986897. 17 Supplementary Materials Fig. S1: Pairwise cosine similarity matrices of bione...

    2020

Pith tools

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