Pith. sign in

REVIEW 4 major objections 6 minor 41 references

NeurOptimisation: The Spiking Way to Evolve

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

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

Pith's one-line read The paper claims that NeurOptimiser, a fully spike-based, asynchronous framework, solves BBOB problems up to 40 dimensions using coordination that emerges from local spiking events rather than a central program, with milliwatt-scale power…

desk verdict Real engineering contribution with public code, but the central claims about full spike-basing, decentralisation, and milliwatt power are ahead of the implemented architecture; re-scope and it is worth engaging seriously. read the letter →

arxiv 2507.08320 v1 pith:WO67PQ2I submitted 2025-07-11 cs.NE

Jorge Mario Cruz-Duarte , El-Ghazali Talbi This is my paper · ORCID
classification cs.NE
keywords NeuromorphicComputingEvent-DrivenAsynchronousOptimisationMetaheuristicsSpikingNeuralNetworksEvolutionaryAlgorithmsDifferentialEvolutionBBOBbenchmarkLoihi2
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

The paper tries to show that a population-based optimiser can be built entirely from spiking neuron events, with no central clock or CPU orchestrator driving the search. Its NeurOptimiser encodes candidate solutions into the internal states of spiking neuron units; when a unit's membrane potential crosses a threshold, the resulting spike triggers a perturbation heuristic, and those spikes propagate to neighbours to share promising positions. The authors implement this in a neuromorphic software framework targeting the Loihi 2 chip and test it on the BBOB benchmark up to 40 dimensions. They report convergence that matches or beats random search in low and moderate dimensions and roughly matches reference optimisers on separable and well-conditioned problems, with milliwatt-level power estimates derived from Loihi event costs.

What carries the argument

The central object is the Neuromorphic Heuristic Unit (NHU): one spiking neuron per coordinate of a candidate solution, each following the generalised transition $v^{t+1} \leftarrow h_d(v^t)$ if $\Phi(v^t, \ldots) \neq 1$ and $v^{t+1} \leftarrow h_s(v^t)$ otherwise, with $\Phi$ the spiking condition, typically $|v_{1,j}| \geq \vartheta_j$. Here $h_d$ supplies the exploration dynamics (linear systems, LIF, or Izhikevich models), $h_s$ acts as the perturbation heuristic (stochastic reset, reset toward a best state, directional displacement, or differential-evolution mutation), and the bidirectional map $T$ translates between problem-space coordinates and neuromorphic states. The mechanism works because the spiking condition is simultaneously a selection predicate and a communication event: a spike both resets or perturbs the unit and propagates through a bulk spike-contraction layer to activate neighbouring units, so search and coordination are the same event.

What would settle it

Run the same NeurOptimiser configurations on a Loihi 2 chip and measure energy per optimisation step; if measured power for the n=90, d=40, m=89 worst case is not in the roughly watt-level range of the estimate, the power-feasibility claim fails. A cheaper check is to instrument the CPU simulation to record total emitted spikes and compare that count with the n(n-1)md synaptic-event formula used in the estimate.

Watch

Extended reading notes

Core claim

On its own terms, the paper's central claim is that the search operator of a metaheuristic can be realised by spiking neuron dynamics rather than by an external program. Each candidate solution is a set of $d$ neurons; their internal states evolve under a dynamic rule $h_d$, and when the spiking condition $\Phi$ triggers, a spike-triggered rule $h_s$ applies a heuristic perturbation, such as a reset toward the best-so-far state or a differential-evolution-style mutation. Local and global coordination, including neighbourhood information, global best sharing, and spike propagation, emerge from asynchronous message passing encoded as spikes rather than from a central loop. The authors argue that this is the first complete and reproducible integration of neuromorphic heuristic-based optimisers, and they report that on BBOB up to 40 dimensions the heterogeneous variants converge reliably on separable, unimodal, and moderately conditioned problems, with runtime linear in population size and dimension, per-unit steps under 12 ms, and power estimates from milliwatts to about 1.35 W in the tested worst case.

Load-bearing premise

The load-bearing premise is that counting events in a CPU simulation and pricing each one with the Loihi chip's published per-event energies gives a realistic picture of real neuromorphic hardware; if real spike counts or per-event costs are substantially different, the milliwatt feasibility claim collapses.

Editorial extensions

If this is right

  • If NeurOptimiser is right, general-purpose optimisation can run on event-driven neuromorphic hardware without a CPU orchestrator, making the same algorithm deployable in low-power embedded settings.
  • The spike-triggered rule is pluggable: both simple resets and differential-evolution mutations are expressible, so established metaheuristic operators can be ported onto spiking substrates rather than re-designed from scratch.
  • Coordination by spike propagation means communication cost scales with actual spiking activity rather than with a fixed synchronisation schedule, so sparse firing directly lowers energy consumption.
  • Heterogeneous populations that mix linear and Izhikevich neuron dynamics proved more robust than homogeneous ones on multimodal and ill-conditioned BBOB problems, pointing to neuron-model diversity as a search resource.
  • Runtime and resource usage scale linearly with population size and dimension, supporting the feasibility of large asynchronous populations on neuromorphic hardware.

Reading between the lines

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

  • Editorial inference: the paper attributes the high-dimensional ECDF plateaus to constrained evaluation budgets rather than algorithmic stagnation; a direct test would run the same variants with budgets comparable to the BBOB 2009 reference and check whether the plateaus lift.
  • The analytical power model assumes worst-case dense connectivity, so a testable extension is to measure actual emitted spikes on the hardware or in simulation and recompute the per-step energy with measured synaptic-event counts, which could bring the estimate below the milliwatt levels already reported.
  • A natural next step the paper leaves implicit is adaptive spike-threshold tuning tied to local convergence, which could reduce superfluous spiking and further lower the energy estimate.
  • The framework defines a design space for choosing $h_d$, $h_s$, and $\Phi$; a promising untested direction is learning which neuron-model or mutation combination suits a given problem class, turning the NHU configuration itself into a search problem.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 6 minor

Summary. The manuscript introduces NeurOptimiser, a framework that combines spiking neuron models with metaheuristic search. A population of Neuromorphic Heuristic Units (NHUs) encodes candidate solutions into neuron states, applies a dynamic rule h_d and a spike-triggered rule h_s (fixed reset or DE/current-to-rand/1), and communicates through a spike matrix, a Neighbour Manager, and a High-Level Selector. The authors implement the system in Intel's Lava framework with Loihi 2 as target, evaluate on noiseless BBOB functions in 2--40 dimensions against BBOB 2009 and RANDOMSEARCH, and estimate power consumption from Loihi per-event costs. They claim this is a fully spike-based, fully decentralised, low-energy optimisation framework.

Significance. The paper is a useful proof of concept: it shows that spike-triggered heuristic rules can drive a population-based search on standard benchmark functions, it ships reproducible code and data, and it benchmarks against external references rather than fitted baselines. The internal-dynamics visualisations (Figs. 6--9) are a strength. However, the significance as claimed in the abstract and conclusion ('fully spike-based', 'fully decentralised', 'milliwatt-level') is not supported by the described implementation, since several coordinating components operate on real-valued position/fitness matrices and all results come from CPU simulation rather than Loihi 2 deployment. The contribution is better framed as a CPU-simulated, spike-triggered heuristic framework with a Loihi target.

major comments (4)
  1. [Section 4.3 (Algorithms 6 and 8)] The claim that coordination arises 'without external orchestration' (Section 7) is contradicted by the architecture. Algorithm 6 collects every NHU's PPP and f_p, performs a global argmin over all units, and broadcasts the real-valued global best ggg; Algorithm 8 centrally builds and redistributes neighbourhood matrices P_n and F_n from all units. These are central coordination processes operating on floating-point data, not native spiking mechanisms.
  2. [Section 4.2 (Algorithms 2, 4, 5) and Eq. (21b)] The 'fully spike-based' claim is not supported: candidate positions are real-valued; the Selector evaluates f on the host CPU and maintains ppp in floating point; Sender/Receiver exchange real-valued position and fitness arrays; and the DE mutation in Eq. (21b) is an arithmetic update on real-valued state vectors, with spikes only determining whether the rule fires. Spikes act as event triggers around CPU heuristics rather than as the substrate of the search, so the strongest formulation in Section 7 ('spiking dynamics ... search engine') is an overstatement.
  3. [Section 6 (Figure 13)] The power feasibility claim is based on an analytical estimate, not a measurement. The formula E_step = 23.6 N_syn + 89.7 n pJ multiplies Loihi per-event costs from [41] by the algorithm's event count and divides by an assumed Delta_t_sim = 0.5 ms; no Loihi 2 chip was used. Moreover, the worst-case estimate reported in Figure 13 is 1.35 W, which contradicts the 'milliwatt-level' wording in the abstract and Section 7. The data support sub-watt estimates for small configurations, but not the stated milliwatt-level feasibility claim.
  4. [Sections 5 and 6] All experiments are CPU simulations; Loihi 2 is only 'targeted'. Statements in Section 6 ('we anticipate further gains') and Section 7 ('suitability ... for practical NC hardware deployments') go beyond the evidence. The authors should confine conclusions to spike-triggered heuristics in simulation and describe the Loihi deployment as future work, or add a measured deployment.
minor comments (6)
  1. [Section 3.3, Eqs. (21a)--(21b)] Two equations are listed but the text says they correspond to three strategies ('current-to-best, rand-to-best, and current-to-rand'); align the number of equations with the description.
  2. [Section 3.2, Eq. (13)] The spike-triggered rule h_s in Eq. (13) appears to return a complex scalar ϑ1 + ϑ2 r1 e^{i2πr2}, while the state space V is real and two-dimensional; clarify the intended construction, for example as a vector in polar coordinates.
  3. [Section 5 vs. Figures 6--9] The first experiment is described as using 1000×d steps, but the captions of Figures 6--9 say 1000 steps; reconcile this inconsistency.
  4. [Section 3.4, Eq. (22)] The random value r introduced in the encoding must be retained for deterministic invertibility; specify where this per-step random value is stored and how T^{-1} accesses it, since the algorithm as written does not keep r in the state.
  5. [Figure 13 caption] There is a typo ('Averate') in the caption of Figure 13, and the caption does not clearly map the parenthetical power values to the (n, d) settings, making the figure harder to read.
  6. [Table 1] The header row for f1 appears garbled ('f111 12 12 ...') in the preprint; ensure the table renders correctly.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity found: the framework is benchmarked against external BBOB baselines and its core claims do not reduce to fitted parameters or to self-citations.

full rationale

The paper's core contribution is an implemented framework whose components are defined independently of the benchmark results. The spiking neuron dynamics (Section 3.2), spike-triggered heuristics (Section 3.3), and coordination processes (Section 4) are all specified through explicit equations and algorithms rather than through fitted parameters. Performance claims are validated against the external BBOB 2009 reference and RANDOMSEARCH baselines using COCO and IOH, and the power feasibility estimate uses per-event costs reported in an external Loihi hardware paper multiplied by the algorithm's own event counts; this is a scaling calculation, not a prediction derived from fitted data. The self-citation [7] supplies the 'Nheuristics' vocabulary and design-space framing, but no experimental result or architectural constraint in this paper is forced by that citation; the paper could stand without it. The only noticeable internal tension is that the High-Level Selector (Algorithm 6) and Neighbour Manager (Algorithm 8) are central, non-spiking coordinators that exchange real-valued position and fitness arrays, which sits uneasily with the abstract's 'without external orchestration' wording. However, this is an internal-consistency or correctness concern about the strength of the claim, not a circularity in the derivation, and it does not make any result equivalent to its inputs by construction. Accordingly, the circularity score remains 0.

Assumptions & free parameters 9 free parameters · 3 assumptions · 2 invented entities

The framework introduces modular abstractions (NHU, Tensor Contraction Layer, Neighbour Manager, High-Level Selector) and relies on several manually set hyperparameters (F, alpha, alpha_thr, n, m, Delta t, weights, sigma, Izhikevich constants). No parameters are fit to the benchmark data in a way that makes the results circular, but many values are not justified in the text.

free parameters (9)
  • F (DE scale factor) = not reported
    In DE/current-to-rand/1 spike-triggered mutation (Eq. 21b); standard DE parameter fixed by the implementer.
  • alpha (encoding gain in T) = not reported in experiments (text suggests alpha=1)
    Scales the mapping between problem coordinate x_j and membrane potential v1,j (Eq. 22); controls search resolution.
  • alpha_thr (threshold base) = not reported
    Sets spike threshold via alpha_thr*|g_j-p_j| (Eq. 16b); a manually fixed scalar governing firing frequency.
  • Delta t (integration step) = 0.01
    Numerical integration step for neuron dynamics (Section 5); fixed for all runs.
  • Izhikevich parameters a,b,c,d = from [31]
    Fixed cortical parameter sets from the literature; not fitted to BBOB.
  • reference weights w1, w2 = 0.5 each
    Weights for x_ref as average of personal and global best (Eq. 23a).
  • number of NHUs, n = 30 (up to 90 in scaling)
    Population size chosen by the authors; affects exploration and runtime.
  • neighbourhood size, m = 10 (up to 89 in scaling)
    Neighbour count for W_x; controls information sharing.
  • sigma (noise std in fixed h_s) = not reported
    Gaussian noise added in fixed reset rule (Eq. 19); set by implementer.
assumptions (3)
  • domain assumption The BBOB noiseless suite is an adequate benchmark for continuous black-box optimization
    Used to claim viability; standard in the field.
  • domain assumption Loihi per-event energy costs from [41] (23.6 pJ per synaptic event, 81 pJ per neuron update, 8.7 pJ per spike) transfer to this Lava simulation workload
    Basis of all power estimates in Figure 13; not measured on this workload.
  • ad hoc to paper Spiking condition (15a) with l2-norm and threshold (16b) interacts with h_d and h_s to produce useful search
    Core design choice, motivated by phase-portrait examples but not proven.
invented entities (2)
  • Neuromorphic Heuristic Unit (NHU)
    purpose: Self-contained module encoding one candidate solution with spiking neurons
    Central architectural abstraction; the paper provides implementation and benchmarks, but no falsifiable prediction outside its own framework.
  • Tensor Contraction Layer
    purpose: Aggregates the spike matrix into activation patterns via a logical AND contraction (Eq. 24)
    A new coordination component; its role is internal to this framework, with no external falsifiable handle.

how reviews work

0 comments
Cite this review

Pith. "Pith review of NeurOptimisation: The Spiking Way to Evolve." pith.science (2026). https://pith.science/paper/WO67PQ2I

@misc{pith2026250708320,
  author       = {Pith},
  title        = {Pith review of: NeurOptimisation: The Spiking Way to Evolve},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/WO67PQ2I}},
  note         = {Machine review of arXiv:2507.08320}
}
read the original abstract

The increasing energy footprint of artificial intelligence systems urges alternative computational models that are both efficient and scalable. Neuromorphic Computing (NC) addresses this challenge by empowering event-driven algorithms that operate with minimal power requirements through biologically inspired spiking dynamics. We present the NeurOptimiser, a fully spike-based optimisation framework that materialises the neuromorphic-based metaheuristic paradigm through a decentralised NC system. The proposed approach comprises a population of Neuromorphic Heuristic Units (NHUs), each combining spiking neuron dynamics with spike-triggered perturbation heuristics to evolve candidate solutions asynchronously. The NeurOptimiser's coordination arises through native spiking mechanisms that support activity propagation, local information sharing, and global state updates without external orchestration. We implement this framework on Intel's Lava platform, targeting the Loihi 2 chip, and evaluate it on the noiseless BBOB suite up to 40 dimensions. We deploy several NeurOptimisers using different configurations, mainly considering dynamic systems such as linear and Izhikevich models for spiking neural dynamics, and fixed and Differential Evolution mutation rules for spike-triggered heuristics. Although these configurations are implemented as a proof of concept, we document and outline further extensions and improvements to the framework implementation. Results show that the proposed approach exhibits structured population dynamics, consistent convergence, and milliwatt-level power feasibility. They also position spike-native MHs as a viable path toward real-time, low-energy, and decentralised optimisation.

Figures

Figures reproduced from arXiv: 2507.08320 by the authors.

Figure 1
Figure 1. Example of three search steps carried out internally by a spiking neuron in an evolutionary process. A [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Phase portraits (v1 vs. v2) and time signals (v1, v2, s vs. t) of the four dynamical behaviours generated using the Izhikevich model and the parameters described in [31]. The dashed black line and orangish-shaded region represent the thresholding condition φs and spiking region S. system experiences an instantaneous reset through hs, as indicated by the abrupt jumps where trajectories are reset to new states. To wra… view at source ↗
Figure 3
Figure 3. Phase portraits (v1 vs. v2) and time signals (v1, v2, s vs. t) of four equilibrium points in a linear system. The dashed black circle and orangish-shaded region depict the thresholding condition φs and spiking region S. The red and blue lines show trajectories with and without spiking, respectively. Foremost, spikes are fired when the neuron’s internal state satisfies a condition φs, which acts as a selection pred￾i… view at source ↗
Figures from the paper (10 more)
Figure 4
Figure 4. Figure 4: Neuromorphic Optimiser (NeurOptimiser) architecture composed of [PITH_FULL_IMAGE:figures/full_fig_p009_4.png]
Figure 5
Figure 5. Figure 5: Neuromorphic Heuristic Unit (NHU) architecture comprising two main processes, Spiking Core and Selec [PITH_FULL_IMAGE:figures/full_fig_p010_5.png]
Figure 6
Figure 6. Figure 6: depicts the evolution of the absolute error εr for these representative 2D problems under certain com￾binations of hd and hs heuristics. We chose to present the most descriptive ones for brevity, although the entire set of results can be consulted on [37]. In both the …
Figure 7
Figure 7. Figure 7: Particular best positions (x1, x2 ∈ p t i ) of a 1000-step simulation with two NeurOptimisers with 30 NHUs searching on 2D problem domains. Each row stands for a function fk from the BBOB suite. Each column corresponds to neurOptimisers (hd+hs) implemented using the Li…
Figure 8
Figure 8. Figure 8: Portraits of neuromorphic state variables ( [PITH_FULL_IMAGE:figures/full_fig_p017_8.png]
Figure 9
Figure 9. Figure 9: Spiking signal count (s1 + s2) of a 1000-step simulation with two NeurOptimisers with 30 NHUs, each one with an index i, searching on 2D problem domains. Each row stands for a function fk from the BBOB suite. Each column corresponds to neurOptimisers (hd+hs) implemente…
Figure 10
Figure 10. Figure 10: Empirical Cumulative Distribution Functions (ECDFs) of normalised runtimes for the three neurOptimiser [PITH_FULL_IMAGE:figures/full_fig_p019_10.png]
Figure 11
Figure 11. Figure 11: Expected Running Time (ERT) loss ratio vs. the budget in number of f-evaluations per dimension. For each budget, the ERT to reach the best target f-value is divided by the ERT of the best BBOB 2009 method for the same target. Lines show the geometric mean; box-whisker…
Figure 12
Figure 12. Figure 12: Bootstrapped Empirical Cumulative Distribution Function (ECDF) of the number of [PITH_FULL_IMAGE:figures/full_fig_p020_12.png]
Figure 13
Figure 13. Figure 13: Average runtime per step and unit (t [ms]) as a function of the number of NHUs n and dimensionality d. Problem instances were evaluated across 2, 10, 20, and 40 dimensions, using configurations with 30, 60, and 90 NHUs. Violin plots display the distribution of per-uni…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

41 extracted references · 37 canonical work pages

  1. [41]

    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, Y . Liao, C.-K. Lin, A. Lines, R. Liu, D. Mathaikutty, S. McCoy, A. Paul, J. Tse, G. Venkataramanan, Y .-H. Weng, A. Wild, Y . Yang, and H. Wang, “Loihi: A neuromorphic manycore processor with on-chip learning,” IEEE Micro, vol. 38, no. 1, p. 82–99,...

    work page 2018

  2. [1]

    An electromagnetic perspec- tive of artificial intelligence neuromorphic chips,

    E.-P. Li, H. Ma, M. Ahmed, T. Tao, Z. Gu, M. Chen, Q. Chen, D. Li, and W. Chen, “An electromagnetic perspec- tive of artificial intelligence neuromorphic chips,”Electromagnetic Science, vol. 1, no. 3, p. 18, 2023

    work page 2023

  3. [2]

    Exploring spiking neural networks: a comprehen- sive analysis of mathematical models and applications,

    Sanaullah, S. Koravuna, U. Rückert, and T. Jungeblut, “Exploring spiking neural networks: a comprehen- sive analysis of mathematical models and applications,”Frontiers in Computational Neuroscience, vol. 17, p. 1215824, 2023

    work page 2023

  4. [3]

    Spiking Neural Networks: A Survey,

    J. D. Nunes, M. Carvalho, D. Carneiro, and J. S. Cardoso, “Spiking Neural Networks: A Survey,”IEEE Access, vol. 10, p. 60738–60764, 2022

    work page 2022

  5. [4]

    Neuromorphic artificial intelligence sys- tems,

    D. Ivanov, A. Chezhegov, M. Kiselev, A. Grunin, and D. Larionov, “Neuromorphic artificial intelligence sys- tems,”Frontiers in Neuroscience, vol. 16, p. 959626, 2022

    work page 2022

  6. [5]

    A survey on evolutionary computation for complex continuous optimization,

    Z.-H. Zhan, L. Shi, K. C. Tan, and J. Zhang, “A survey on evolutionary computation for complex continuous optimization,”Artificial Intelligence Review, vol. 55, no. 1, pp. 59–110, 2022

    work page 2022

  7. [6]

    Large-scale evolutionary optimization: A review and comparative study,

    J. Liu, R. Sarker, S. Elsayed, D. Essam, and N. Siswanto, “Large-scale evolutionary optimization: A review and comparative study,”Swarm and Evolutionary Computation, vol. 85, p. 101466, 2024

    work page 2024

  8. [7]

    Neuromorphic-based metaheuristics: A new generation of low power, low latency and small footprint optimization algorithms

    E. Talbi, “Neuromorphic-based metaheuristics: A new generation of low power, low latency and small footprint optimization algorithms,” 2025. [Online]. Available: https://arxiv.org/abs/2505.16362

    work page Pith review arXiv 2025

Show all 41 references
  1. [8]

    Lava: A software framework for neuromorphic computing,

    Intel Neuromorphic Computing Lab, “Lava: A software framework for neuromorphic computing,” https://github. com/lava-nc, 2023, accessed: 2025-03-28

    2023

  2. [9]

    COCO: A platform for comparing continuous optimizers in a black-box setting,

    N. Hansen, A. Auger, R. Ros, O. Mersmann, T. Tušar, and D. Brockhoff, “COCO: A platform for comparing continuous optimizers in a black-box setting,”Optimization Methods and Software, vol. 36, pp. 114–144, 2021

    2021

  3. [10]

    A ferroelectric memristor-based transient chaotic neural network for solving combinatorial optimization problems,

    Z. Lin and Z. Fan, “A ferroelectric memristor-based transient chaotic neural network for solving combinatorial optimization problems,”Symmetry, vol. 15, no. 1, p. 59, 2022

    2022

  4. [11]

    Organic memristor-based flexible neural networks with bio-realistic synaptic plasticity for complex combinatorial optimization,

    H. Kim, M. Kim, A. Lee, H. Park, J. Jang, J. Bae, I. M. Kang, E. Kim, and S. Lee, “Organic memristor-based flexible neural networks with bio-realistic synaptic plasticity for complex combinatorial optimization,”Advanced Science, vol. 10, no. 19, p. 2300659, 2023

    2023

  5. [12]

    Neuromorphic swarm on RRAM compute-in-memory processor for solving QUBO problem,

    A. S. Lele, M. Chang, S. D. Spetalnick, B. Crafton, A. Raychowdhury, and Y . Fang, “Neuromorphic swarm on RRAM compute-in-memory processor for solving QUBO problem,”2023 60th ACM/IEEE Design Automation Conference (DAC), vol. 00, pp. 1–6, 2023

    2023

  6. [13]

    Solving QUBO on the Loihi 2 Neuromorphic Processor,

    A. Pierro, P. Stratmann, G. A. F. Guerra, S. Risbud, T. Shea, A. R. Mangalore, and A. Wild, “Solving QUBO on the Loihi 2 Neuromorphic Processor,”arXiv, 2024

    2024

  7. [14]

    Towards spike-based machine intelligence with neuromorphic computing,

    K. Roy, A. Jaiswal, and P. Panda, “Towards spike-based machine intelligence with neuromorphic computing,” Nature, vol. 575, no. 7784, pp. 607–617, 2019

    2019

  8. [15]

    Evolutionary spiking neural networks: A survey,

    S. Shen, R. Zhang, C. Wang, R. Huang, A. Tuerhong, Q. Guo, Z. Lu, J. Zhang, and L. Leng, “Evolutionary spiking neural networks: A survey,”arXiv, 2024. [Online]. Available: 10.48550/arxiv.2406.12552

    2024 doi arXiv

  9. [16]

    Evolutionary optimization for neuromorphic systems,

    C. D. Schuman, J. P. Mitchell, R. M. Patton, T. E. Potok, and J. S. Plank, “Evolutionary optimization for neuromorphic systems,” inProceedings of the 2020 Annual Neuro-Inspired Computational Elements Workshop, ser. NICE’20. New York, NY , USA: ACM, 2020. [Online]. Available: h...

    2020

  10. [17]

    Real-time evolution and deployment of neu- romorphic computing at the edge,

    C. D. Schuman, S. R. Young, B. P. Maldonado, and B. C. Kaul, “Real-time evolution and deployment of neu- romorphic computing at the edge,” in2021 12th International Green and Sustainable Computing Conference (IGSC). IEEE, 2021, pp. 1–8

    2021

  11. [18]

    A swarm optimization solver based on ferroelectric spiking neural networks,

    Y . Fang, Z. Wang, J. Gomez, S. Datta, A. I. Khan, and A. Raychowdhury, “A swarm optimization solver based on ferroelectric spiking neural networks,”Frontiers in Neuroscience, vol. 13, p. 855, 2019

    2019

  12. [19]

    Swarm intelligence algorithm based on spiking neural-oscillator networks, coupling interactions and search performances,

    T. Sasaki and H. Nakano, “Swarm intelligence algorithm based on spiking neural-oscillator networks, coupling interactions and search performances,”Nonlinear Theory and Its Applications, IEICE, vol. 14, no. 2, p. 267–291, 2023. 23 NeurOptimisation: The Spiking Way to EvolvePREPRINT

    2023

  13. [20]

    Advancements in algorithms and neuromorphic hard- ware for spiking neural networks,

    A. Javanshir, T. Nguyen, M. Mahmud, and A. Kouzani, “Advancements in algorithms and neuromorphic hard- ware for spiking neural networks,”Neural Computation, vol. 34, no. 6, p. 1289–1328, 2022

    2022

  14. [21]

    Neuromorphic bayesian optimization in Lava,

    S. Snyder, S. R. Risbud, and M. Parsa, “Neuromorphic bayesian optimization in Lava,” inProc. International Conference on Neuromorphic Systems, ser. ICONS ’23. New York, NY , USA: ACM, 2023. [Online]. Available: https://doi.org/10.1145/3589737.3605998

    2023

  15. [22]

    Asynchronous neuromorphic optimization in Lava,

    S. Snyder, S. Risbud, and M. Parsa, “Asynchronous neuromorphic optimization in Lava,” inProc. Great Lakes Symposium on VLSI 2024, ser. GLSVLSI ’24. New York, NY , USA: ACM, 2024, p. 776–778. [Online]. Available: https://doi.org/10.1145/3649476.3660383

    2024

  16. [23]

    Parallelized multi-agent bayesian optimization in Lava,

    S. Snyder, D. Gobin, V . Clerico, S. R. Risbud, and M. Parsa, “Parallelized multi-agent bayesian optimization in Lava,” inProc. International Conference on Neuromorphic Systems, 2024, pp. 295–298

    2024

  17. [24]

    Hyper-heuristics to customise metaheuristics for continuous optimisation,

    J. Cruz-Duarte, I. Amaya, J. Ortiz-Bayliss, S. Conant-Pablos, H. Terashima-Marín, and Y . Shi, “Hyper-heuristics to customise metaheuristics for continuous optimisation,”Swarm Evol. Comput., vol. 66, p. 100935, 2021

    2021

  18. [25]

    Recent advances in selection hyper-heuristics,

    J. H. Drake, A. Kheiri, E. Özcan, and E. K. Burke, “Recent advances in selection hyper-heuristics,”European Journal of Operational Research, vol. 285, pp. 405–428, 2020, https://linkinghub.elsevier.com/retrieve/pii/S0377221719306526. [Online]. Available: https://doi.org/10.101...

    2020 doi

  19. [26]

    Talbi,Metaheuristics: from design to implementation

    E.-G. Talbi,Metaheuristics: from design to implementation. John Wiley & Sons, 2009

    2009

  20. [27]

    Towards a generalised metaheuristic model for continuous optimisation problems,

    J. M. Cruz-Duarte, J. C. Ortiz-Bayliss, I. Amaya, Y . Shi, H. Terashima-Marín, and N. Pillay, “Towards a generalised metaheuristic model for continuous optimisation problems,”Mathematics, vol. 8, pp. 1–23, 11 2020. [Online]. Available: https://www.mdpi.com/2227-7390/8/11/2046

    2020

  21. [28]

    Differential evolution mutations: Taxonomy, comparison and convergence analysis,

    A. W. Mohamed, A. A. Hadi, and A. K. Mohamed, “Differential evolution mutations: Taxonomy, comparison and convergence analysis,”IEEE Access, vol. 9, pp. 68 629–68 662, 2021

    2021

  22. [29]

    Spiking neural networks and their applications: A review,

    K. Yamazaki, V .-K. V o-Ho, D. Bulsara, and N. Le, “Spiking neural networks and their applications: A review,” Brain Sciences, vol. 12, no. 7, p. 863, 2022

    2022

  23. [30]

    Opportunities for neuromorphic computing algorithms and applications,

    C. D. Schuman, S. R. Kulkarni, M. Parsa, J. P. Mitchell, P. Date, and B. Kay, “Opportunities for neuromorphic computing algorithms and applications,”Nature Computational Science, vol. 2, no. 1, p. 10–19, 2022

    2022

  24. [31]

    Simple model of spiking neurons,

    E. M. Izhikevich, “Simple model of spiking neurons,”IEEE Transactions on Neural Networks, vol. 14, no. 6, p. 1569, 2003

    2003

  25. [32]

    H. S. Steven,Nonlinear dynamics and chaos: with applications to physics, biology, chemistry, and engineering. CHAPMAN & HALL CRC, 2024

    2024

  26. [33]

    Tornado: An Autonomous Chaotic Algorithm for Large Scale Global Optimization,

    N. Aslimani, E.-G. Talbi, and R. Ellaia, “Tornado: An Autonomous Chaotic Algorithm for Large Scale Global Optimization,” Mar. 2020, working paper or preprint. [Online]. Available: https://inria.hal.science/hal-02499326

    2020

  27. [34]

    A review of computational models for gamma oscillation dynamics: from spiking neurons to neural masses,

    Z. Liu, F. Han, and Q. Wang, “A review of computational models for gamma oscillation dynamics: from spiking neurons to neural masses,”Nonlinear Dynamics, vol. 108, no. 3, p. 1849–1866, 2022

    2022

  28. [35]

    Spiking neural networks for nonlinear regression,

    A. Henkes, J. K. Eshraghian, and H. Wessels, “Spiking neural networks for nonlinear regression,”Royal Society Open Science, vol. 11, no. 5, p. 231606, 2024

    2024

  29. [36]

    Tensor Contraction Layers for Parsimo- nious Deep Nets,

    J. Kossaifi, A. Khanna, Z. Lipton, T. Furlanello, and A. Anandkumar, “Tensor Contraction Layers for Parsimo- nious Deep Nets,”2017 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), pp. 1940–1946, 2017

    2017

  30. [37]

    Neuroptimisation: The spiking way to evolve - experiment codes and dataset,

    J. M. Cruz-Duarte and E.-G. Talbi, “Neuroptimisation: The spiking way to evolve - experiment codes and dataset,” Jul. 2025. [Online]. Available: https://doi.org/10.5281/zenodo.15858610

    2025 doi

  31. [38]

    Real-parameter black-box optimization benchmarking 2009: Noiseless functions definitions,

    S. Finck, N. Hansen, R. Ros, and A. Auger, “Real-parameter black-box optimization benchmarking 2009: Noiseless functions definitions,” INRIA, Tech. Rep. RR-6829, 2009, updated version as of Feb. 2019. [Online]. Available: https://inria.hal.science/inria-00362633v2

    2009

  32. [39]

    IOHexperimenter: Benchmarking Platform for Iterative Optimization Heuristics,

    J. de Nobel, F. Ye, D. Vermetten, H. Wang, C. Doerr, and T. Bäck, “IOHexperimenter: Benchmarking Platform for Iterative Optimization Heuristics,”arXiv, Nov. 2021. [Online]. Available: https://arxiv.org/abs/2111.04077

    2021 arXiv

  33. [40]

    Benchmarking the pure random search on the bbob-2009 testbed,

    A. Auger and R. Ros, “Benchmarking the pure random search on the bbob-2009 testbed,” inProc. 11th Annual Conference Companion on Genetic and Evolutionary Computation Conference: Late Breaking Papers, 2009, pp. 2479–2484

    2009

Pith tools

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