pith. machine review for the scientific record. sign in

arxiv: 2605.01821 · v1 · submitted 2026-05-03 · ⚛️ physics.soc-ph

Recognition: unknown

Adjacent Possible Innovation Dynamics on Local Optima Networks

Edward D. Lee, J\'anos Kert\'esz, Leonardo Rizzo

Authors on Pith no claims yet

Pith reviewed 2026-05-09 16:00 UTC · model grok-4.3

classification ⚛️ physics.soc-ph
keywords innovation dynamicslocal optima networksHeaps' lawZipf's lawTaylor's lawadjacent possibletechnological paradigmsfitness landscapes
0
0 comments X

The pith

Modeling innovation as random walks on networks of stable technological configurations reproduces four key empirical patterns of discovery.

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

The paper proposes that innovation can be represented as agents walking across a Local Optima Network, a graph whose nodes are locally optimal technology setups and whose edges capture the probabilities of moving from one setup to another. This single representation is shown to generate sublinear growth in new ideas, heavy-tailed frequencies of use, specific scaling of fluctuations, and power-law waiting times between discoveries, all with exponents that fall inside the ranges seen in real data. Because the network structure alone sets these exponents, the approach unifies two traditions in innovation studies: one focused on statistical regularities of discovery and the other on search within a space of possibilities. The result is a compact, testable model that treats technological paradigms as communities of mutually accessible configurations.

Core claim

Local Optima Networks constructed from fitness landscapes allow stochastic walkers to explore the adjacent possible in a way that simultaneously produces Heaps' law for novelty growth, Zipf's law for frequency distributions, Taylor's law for fluctuation scaling, and power-law inter-event times, with all four exponents jointly fixed by the network topology and lying in empirically observed ranges.

What carries the argument

Local Optima Network: a directed weighted graph in which nodes stand for basins of attraction around local fitness maxima and edges encode transition probabilities between basins; agents perform stochastic walks on this graph to model discovery.

If this is right

  • Communities within the network supply an operational definition of technological paradigms based on shared basin accessibility.
  • The same framework can represent both the statistical signatures of discovery and the mechanics of adaptive search.
  • Exponents are constrained jointly rather than independently, so changing network structure affects all four laws in linked ways.
  • The model is parsimonious because it requires no separate mechanisms for each empirical regularity.

Where Pith is reading between the lines

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

  • If real innovation landscapes yield Local Optima Networks with measurable community structure, the model predicts that paradigm shifts should correspond to detectable jumps between those communities.
  • The framework suggests that data on patent or product transitions could be used to infer the underlying fitness landscape and then test the predicted exponents directly.
  • Extensions to time-varying networks could allow the model to capture how technological landscapes themselves evolve.

Load-bearing premise

The topology of a Local Optima Network built from a fitness landscape is by itself enough to force the four exponents into their observed ranges without extra parameter tuning or special choice of landscape.

What would settle it

Construct a Local Optima Network from an empirically measured technological fitness landscape, run the walker model, and check whether the generated exponents for the four laws fall outside the ranges measured in corresponding real innovation data.

Figures

Figures reproduced from arXiv: 2605.01821 by Edward D. Lee, J\'anos Kert\'esz, Leonardo Rizzo.

Figure 1
Figure 1. Figure 1: Fitness landscapes varying ωL (columns, increasing left to right) and persistence ρ (rows, increasing top to bottom), with L = 1000 and n = 6 octaves. The number of local optima identified by exhaustive hill climbing is shown in each panel. Fitness is normalized to [0, 100]. 3.3 Local Optima Network Construction 3.3.1 Network definition Definition 8 (Local Optima Network). A Local Optima Network is a direc… view at source ↗
Figure 2
Figure 2. Figure 2: Left: a representative fitness landscape with local optima marked as stars. view at source ↗
Figure 3
Figure 3. Figure 3: Node-level attribute relationships in the baseline LON ( view at source ↗
Figure 4
Figure 4. Figure 4: Community structure of a representative LON. (a) Fitness landscape with view at source ↗
Figure 5
Figure 5. Figure 5: Punctuated discovery dynamics. A single walk on the baseline LON (|V | = 47,272 nodes). (a) Fitness change ∆f = fnew − fprev at each discovery event. Red (blue) spikes indicate fitness improvements (regressions); events cluster into bursts separated by quiescent intervals. (b) Cumulative distinct optima D(t). Plateaus correspond to trapping episodes; risers correspond to burst phases. This punctuated patte… view at source ↗
Figure 6
Figure 6. Figure 6: Baseline ensemble results (ρ = 0.8, n = 7 octaves, ε = 0; 20 LONs × 50 walks = 1,000 replications, T = 200,000). (a) Heaps’ law: distinct optima D(t) versus step t, with ensemble mean, percentile bands, and fitted exponent β = 0.763. (b) Taylor’s law: σ[D(t)] versus E[D(t)], with fitted exponent b = 1.015. (c) Zipf’s law: rank-frequency distribution (representative run), with tail fit α = 1.148 for ranks R… view at source ↗
Figure 7
Figure 7. Figure 7: Ensemble results with reduced landscape complexity ( view at source ↗
Figure 8
Figure 8. Figure 8: Ensemble results with higher persistence ( view at source ↗
Figure 9
Figure 9. Figure 9: Ensemble results with exogenous innovation ( view at source ↗
read the original abstract

We propose Local Optima Networks (LONs) as a formal framework for modeling innovation dynamics. A LON is a directed weighted graph in which nodes represent locally stable technological configurations and edges encode transition probabilities between their basins of attraction. We construct LONs from fitness landscapes and model innovating agents as stochastic walkers exploring the adjacent possible on the resulting network. We show that this model simultaneously generates the four main empirical regularities of the discovery-process tradition: sublinear novelty growth (Heaps' law), heavy-tailed frequency distributions (Zipf's law), anomalous fluctuation scaling (Taylor's law), and power-law distributed inter-event times. The exponents fall within empirically observed ranges and are jointly constrained by LON topology. Communities in the LON provide an operational definition of technological paradigms grounded in basin-level accessibility. The LON framework thus bridges the discovery-process and adaptive-search traditions of innovation modeling within a single, parsimonious, and empirically testable representation.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

3 major / 2 minor

Summary. The manuscript proposes Local Optima Networks (LONs) as a framework for innovation dynamics, constructing directed weighted graphs from fitness landscapes where nodes are locally optimal technological configurations and edges encode basin transition probabilities. Stochastic walkers on these networks are claimed to simultaneously reproduce four empirical regularities—sublinear novelty growth (Heaps' law), heavy-tailed frequencies (Zipf's law), anomalous fluctuation scaling (Taylor's law), and power-law inter-event times—with exponents in observed ranges jointly fixed by LON topology. Communities in the LON are offered as an operational definition of technological paradigms, unifying discovery-process and adaptive-search modeling traditions in a parsimonious representation.

Significance. If substantiated, the work would offer a concrete network-based bridge between two innovation-modeling traditions by showing how landscape-derived topology can constrain multiple statistical signatures without separate mechanisms. The explicit linkage of basin accessibility to paradigm-like communities and the potential for falsifiable predictions from measurable LON features would be a notable contribution to complex-systems approaches in innovation studies.

major comments (3)
  1. [Abstract, §2] Abstract and §2 (Model Construction): The procedure for building LONs from fitness landscapes is stated at a high level only; no specific landscape family (e.g., NK-model with epistasis parameter K), basin-computation algorithm, or transition-probability estimation method is supplied. Without this, it is impossible to determine whether the topology is generated independently of the target innovation statistics or whether landscape parameters are chosen post hoc to place the four exponents inside empirical ranges.
  2. [§4] §4 (Results): The claim that the same LON topology jointly constrains the exponents of Heaps', Zipf's, Taylor's, and inter-event laws lacks any analytical derivation or sensitivity analysis linking concrete topological quantities (basin-size distribution, edge-weight heterogeneity, community structure) to those exponents. Only simulation outcomes are referenced; the manuscript provides neither the walker update rule parameters, the number of independent landscape realizations, nor the fitting procedure used to extract the four scaling relations.
  3. [§5] §5 (Empirical Comparison): The assertion that simulated exponents fall within observed ranges is not accompanied by quantitative comparisons (e.g., Kolmogorov-Smirnov statistics, R² values, or bootstrap confidence intervals) against specific innovation datasets. Consequently, the statement that the exponents are “jointly constrained by LON topology” rather than by free parameters or selective landscape choice remains unverified.
minor comments (2)
  1. [§3] Notation for the stochastic walker transition probabilities is introduced without an explicit equation; adding a numbered equation in §3 would improve traceability when the results are later discussed.
  2. [Figures] Figure captions for the LON schematics and community visualizations should explicitly state what node size, edge thickness, and color represent (basin volume, transition probability, community label).

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the detailed and constructive comments, which highlight important areas for clarification and strengthening of the methodological and empirical sections. We will revise the manuscript to address these points directly.

read point-by-point responses

  1. Referee: [Abstract, §2] Abstract and §2 (Model Construction): The procedure for building LONs from fitness landscapes is stated at a high level only; no specific landscape family (e.g., NK-model with epistasis parameter K), basin-computation algorithm, or transition-probability estimation method is supplied. Without this, it is impossible to determine whether the topology is generated independently of the target innovation statistics or whether landscape parameters are chosen post hoc to place the four exponents inside empirical ranges.

    Authors: We agree that the construction details were presented at a high level in the submitted version. In the revised manuscript we will expand §2 with the explicit landscape family (NK model with K=2, selected from prior innovation-landscape literature), the basin-computation procedure (exhaustive enumeration for the small landscapes used), and the transition-probability estimation (via 10^4 random walks per basin). We will also add a robustness check across a narrow range of K values to show that the reported topological features and resulting exponents are not the result of post-hoc tuning. revision: yes

  2. Referee: [§4] §4 (Results): The claim that the same LON topology jointly constrains the exponents of Heaps', Zipf's, Taylor's, and inter-event laws lacks any analytical derivation or sensitivity analysis linking concrete topological quantities (basin-size distribution, edge-weight heterogeneity, community structure) to those exponents. Only simulation outcomes are referenced; the manuscript provides neither the walker update rule parameters, the number of independent landscape realizations, nor the fitting procedure used to extract the four scaling relations.

    Authors: The referee correctly notes the absence of an analytical derivation and the limited simulation details. The work is simulation-driven; we will revise §4 to specify the walker rule (memoryless random walk using the directed edge weights), the number of independent realizations (50 per parameter combination), and the fitting protocol (maximum-likelihood power-law estimation with KS goodness-of-fit). We will further include a sensitivity panel relating basin-size heterogeneity and community modularity to the four exponents. revision: yes

  3. Referee: [§5] §5 (Empirical Comparison): The assertion that simulated exponents fall within observed ranges is not accompanied by quantitative comparisons (e.g., Kolmogorov-Smirnov statistics, R² values, or bootstrap confidence intervals) against specific innovation datasets. Consequently, the statement that the exponents are “jointly constrained by LON topology” rather than by free parameters or selective landscape choice remains unverified.

    Authors: We accept that the original text lacked formal statistical comparisons. The revision will add quantitative matches to concrete datasets (patent and citation records for Zipf and Heaps exponents, product-category data for Taylor’s law and inter-event times), reporting KS statistics, R² values, and bootstrap confidence intervals. These additions will allow readers to assess whether the joint reproduction arises from the shared LON topology. revision: yes

Circularity Check

0 steps flagged

No significant circularity detected; derivation remains self-contained.

full rationale

The paper constructs LONs from independent fitness landscapes, then deploys stochastic walkers on the resulting directed weighted graph to generate the four empirical laws (Heaps, Zipf, Taylor, inter-event). No quoted step defines the network topology or transition probabilities in terms of the target exponents or frequency distributions; the landscapes are presented as an external input whose topological features (basin sizes, edge weights) are claimed to jointly constrain the outputs without post-hoc fitting to the innovation data. Because no equation or construction procedure reduces the predicted statistics to the inputs by definition or by self-citation load-bearing, the central claim does not collapse into a tautology. This is the expected non-finding when the derivation chain is not shown to be closed on itself.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The framework rests on the existence of fitness landscapes that can be turned into directed weighted graphs whose basin structure produces the observed laws when traversed by memoryless walkers. No explicit free parameters, axioms, or invented entities are named in the abstract.

axioms (2)
  • domain assumption Fitness landscapes exist and can be mapped to Local Optima Networks with well-defined basins of attraction.
    Invoked when the paper states 'We construct LONs from fitness landscapes'.
  • domain assumption Transition probabilities between basins are sufficient to model agent movement without memory or global knowledge.
    Stated by modeling agents as 'stochastic walkers exploring the adjacent possible'.

pith-pipeline@v0.9.0 · 5458 in / 1395 out tokens · 26147 ms · 2026-05-09T16:00:46.030557+00:00 · methodology

discussion (0)

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

Reference graph

Works this paper leans on

89 extracted references · 1 canonical work pages

  1. [1]

    Herdan, Gustav , title =

  2. [2]

    , title =

    Heaps, Harold S. , title =

  3. [3]

    , title =

    Zipf, George K. , title =

  4. [4]

    ZAMM -- Journal of Applied Mathematics and Mechanics , volume =

    Eggenberger, Felix and P. ZAMM -- Journal of Applied Mathematics and Mechanics , volume =

  5. [5]

    Sur quelques points de la th

    P. Sur quelques points de la th. Annales de l'Institut Henri Poincar

  6. [6]

    Udny , title =

    Yule, G. Udny , title =. Philosophical Transactions of the Royal Society of London. Series B , volume =

  7. [7]

    , title =

    Simon, Herbert A. , title =. Biometrika , volume =

  8. [8]

    , title =

    Hoppe, Fred M. , title =. Journal of Mathematical Biology , volume =

  9. [9]

    , title =

    Ewens, Warren J. , title =. Theoretical Population Biology , volume =

  10. [10]

    Tria, Francesca and Loreto, Vittorio and Servedio, Vito D. P. and Strogatz, Steven H. , title =. Scientific Reports , volume =

  11. [11]

    Loreto, Vittorio and Servedio, Vito D. P. and Strogatz, Steven H. and Tria, Francesca , title =. Creativity and Universality in Language , editor =. 2016 , isbn =

    2016

  12. [12]

    Tria, Francesca and Loreto, Vittorio and Servedio, Vito D. P. , title =. Entropy , volume =

  13. [13]

    Tria, Francesca and Crimaldi, Irene and Aletti, Giovanni and Servedio, Vito D. P. , title =. Entropy , volume =. 2020 , doi =

    2020

  14. [14]

    and Ruiz-Serrano,

    Monechi, B. and Ruiz-Serrano,. Waves of novelties in the expansion into the adjacent possible , journal =

  15. [15]

    and Milojevi

    Iacopini, I. and Milojevi. Network dynamics of innovation processes , journal =

  16. [16]

    and Burioni, R

    Ubaldi, E. and Burioni, R. and Loreto, V. and Tria, F. , title =. Communications Physics , volume =

  17. [17]

    Scientific Reports , volume=

    Modeling innovation in the cryptocurrency ecosystem , author=. Scientific Reports , volume=. 2022 , publisher=

    2022

  18. [18]

    de Solla , title =

    Price, Derek J. de Solla , title =. Science , volume =

  19. [19]

    Emergence of scaling in random networks , journal =

    Barab. Emergence of scaling in random networks , journal =

  20. [20]

    Physical Review Letters , volume =

    Bak, Per and Sneppen, Kim , title =. Physical Review Letters , volume =

  21. [21]

    New Journal of Physics , volume =

    Thurner, Stefan and Klimek, Peter and Hanel, Rudolf , title =. New Journal of Physics , volume =

  22. [22]

    , title =

    Schumpeter, Joseph A. , title =

  23. [23]

    , title =

    Kauffman, Stuart A. , title =

  24. [24]

    Research Policy , year =

    Dosi, Giovanni , title =. Research Policy , year =

  25. [25]

    Management Science , year =

    Fleming, Lee , title =. Management Science , year =

  26. [26]

    Economics of Innovation and New Technology , year =

    Frenken, Koen , title =. Economics of Innovation and New Technology , year =

  27. [27]

    Management Science , year =

    Gavetti, Giovanni and Levinthal, Daniel , title =. Management Science , year =

  28. [28]

    , title =

    Levinthal, Daniel A. , title =. Management Science , year =

  29. [29]

    , title =

    March, James G. , title =. Organization Science , year =

  30. [30]

    and Winter, Sidney G

    Nelson, Richard R. and Winter, Sidney G. , title =

  31. [31]

    A study of

    Ochoa, Gabriela and Tomassini, Marco and V. A study of. Proceedings of the 10th Annual Conference on Genetic and Evolutionary Computation (GECCO) , year =

  32. [32]

    ACM SIGGRAPH Computer Graphics , year =

    Perlin, Ken , title =. ACM SIGGRAPH Computer Graphics , year =

  33. [33]

    and Koput, Kenneth W

    Powell, Walter W. and Koput, Kenneth W. and Smith-Doerr, Laurel , title =. Administrative Science Quarterly , year =

  34. [34]

    and Foster, Ian T

    Rzhetsky, Andrey and Foster, Jacob G. and Foster, Ian T. and Evans, James A. , title =. Proceedings of the National Academy of Sciences , year =

  35. [35]

    Local optima networks of

    V. Local optima networks of. IEEE Transactions on Evolutionary Computation , year =

  36. [36]

    Proceedings of the Sixth International Congress of Genetics , year =

    Wright, Sewall , title =. Proceedings of the Sixth International Congress of Genetics , year =

  37. [37]

    Structural Change and Economic Dynamics , volume =

    Complex economic problems and fitness landscapes: Assessment and methodological perspectives , author =. Structural Change and Economic Dynamics , volume =. 2020 , doi =

    2020

  38. [38]

    Recent Advances in Intelligent Engineering Systems , publisher =

    A Comprehensive Survey on Fitness Landscape Analysis , author =. Recent Advances in Intelligent Engineering Systems , publisher =

  39. [39]

    Evolutionary Computation , volume =

    Information characteristics and the structure of landscapes , author =. Evolutionary Computation , volume =

  40. [40]

    Journal of Statistical Mechanics: Theory and Experiment , volume =

    Quantitative analyses of empirical fitness landscapes , author =. Journal of Statistical Mechanics: Theory and Experiment , volume =

  41. [41]

    Nature Reviews Genetics , volume =

    Empirical fitness landscapes and the predictability of evolution , author =. Nature Reviews Genetics , volume =

  42. [42]

    Research Policy , volume =

    Mapping the technological landscape: Measuring technology distance, technological footprints, and technology evolution , author =. Research Policy , volume =

  43. [43]

    Research Policy , volume =

    Technology as a complex adaptive system: evidence from patent data , author =. Research Policy , volume =

  44. [44]

    International Journal of Operations & Production Management , volume =

    Manufacturing strategy: understanding the fitness landscape , author =. International Journal of Operations & Production Management , volume =

  45. [45]

    Proceedings of the National Academy of Sciences , volume =

    Role of design complexity in technology improvement , author =. Proceedings of the National Academy of Sciences , volume =

  46. [46]

    Journal of Theoretical Biology , volume =

    Towards a general theory of adaptive walks on rugged landscapes , author =. Journal of Theoretical Biology , volume =

  47. [47]

    Artificial Life , volume =

    NK fitness landscapes , author =. Artificial Life , volume =

  48. [48]

    Management Science , volume =

    Modularity and innovation in complex systems , author =. Management Science , volume =

  49. [49]

    Journal of Theoretical Biology , volume =

    Coevolution to the edge of chaos: coupled fitness landscapes, poised states, and coevolutionary avalanches , author =. Journal of Theoretical Biology , volume =

  50. [50]

    The Review of Economic Studies , volume =

    Nonergodic economic growth , author =. The Review of Economic Studies , volume =

  51. [51]

    Journal of Economic Behavior & Organization , volume =

    Optimal search on a technology landscape , author =. Journal of Economic Behavior & Organization , volume =

  52. [52]

    Strategic Entrepreneurship Journal , volume =

    Economic opportunity and evolution: beyond landscapes and bounded rationality , author =. Strategic Entrepreneurship Journal , volume =

  53. [53]

    Organization Science , volume =

    Landscape design: designing for local action in complex worlds , author =. Organization Science , volume =

  54. [54]

    Administrative Science Quarterly , volume =

    Looking forward and looking backward: cognitive and experiential search , author =. Administrative Science Quarterly , volume =

  55. [55]

    Proceedings of the National Academy of Sciences , volume =

    Collaborative learning in networks , author =. Proceedings of the National Academy of Sciences , volume =

  56. [56]

    Local Optima Networks: A New Model of Combinatorial Fitness Landscapes

    Ochoa, Gabriela and Verel, S \'e bastien and Daolio, Fabio and Tomassini, Marco. Local Optima Networks: A New Model of Combinatorial Fitness Landscapes. Recent Advances in the Theory and Application of Fitness Landscapes. 2014

    2014

  57. [57]

    A study of

    Ochoa, Gabriela and Tomassini, Marco and Verel, S. A study of. Proceedings of the Genetic and Evolutionary Computation Conference (GECCO 2008) , pages =

    2008

  58. [58]

    Parallel Problem Solving from Nature -- PPSN XII , series =

    Local optima networks, landscape autocorrelation and heuristic search performance , author =. Parallel Problem Solving from Nature -- PPSN XII , series =

  59. [59]

    Science , volume =

    A topographic view of supercooled liquids and glass formation , author =. Science , volume =

  60. [60]

    Physical Review Letters , volume =

    The network topology of a potential energy landscape: a static scale-free network , author =. Physical Review Letters , volume =

  61. [61]

    Physical Review E , volume =

    Identifying communities within energy landscapes , author =. Physical Review E , volume =

  62. [62]

    Natural Language Engineering , volume =

    Extending Zipf's law to n -grams for large corpora , author =. Natural Language Engineering , volume =

  63. [63]

    Scientific Reports , volume =

    Zipf's law holds for phrases, not words , author =. Scientific Reports , volume =

  64. [64]

    On the evolution of random graphs , journal =

    Erd. On the evolution of random graphs , journal =

  65. [65]

    Journal of Graph Algorithms and Applications , volume =

    Pons, Pascal and Latapy, Matthieu , title =. Journal of Graph Algorithms and Applications , volume =

  66. [66]

    Stauffer, Dietrich and Aharony, Amnon , title =

  67. [67]

    and Ochoa, Gabriela , title =

    Cleghorn, Christopher W. and Ochoa, Gabriela , title =. Proceedings of the 2021 Genetic and Evolutionary Computation Conference Companion , pages =. 2021 , doi =

    2021

  68. [68]

    and Pappa, Gisele L

    Teixeira, Matheus C. and Pappa, Gisele L. , title =. Proceedings of the Genetic and Evolutionary Computation Conference , pages =. 2022 , doi =

    2022

  69. [69]

    and Ochoa, Gabriela and Engelbrecht, Andries P

    Mostert, Werner and Malan, Katherine M. and Ochoa, Gabriela and Engelbrecht, Andries P. , title =. Evolutionary Computation in Combinatorial Optimization , editor =. 2019 , doi =

    2019

  70. [70]

    and Ochoa, Gabriela and Verel, S

    Thomson, Sarah L. and Ochoa, Gabriela and Verel, S. Inferring Future Landscapes: Sampling the Local Optima Level , journal =. 2020 , doi =

    2020

  71. [71]

    Memetic Differential Evolution Using Network Centrality Measures , journal =

    Homolya, Viktor and Vink. Memetic Differential Evolution Using Network Centrality Measures , journal =. 2019 , doi =

    2019

  72. [72]

    Leveraging Local Optima Network Properties for Memetic Differential Evolution , booktitle =

    Homolya, Viktor and Vink. Leveraging Local Optima Network Properties for Memetic Differential Evolution , booktitle =. 2020 , doi =

    2020

  73. [73]

    and Le Goff, L

    Thomson, Sarah L. and Le Goff, L. Understanding Fitness Landscapes in Morpho-Evolution via Local Optima Networks , booktitle =. 2024 , doi =

    2024

  74. [74]

    , title =

    Blackwell, David and MacQueen, James B. , title =. The Annals of Statistics , volume =

  75. [75]

    The Annals of Probability , volume =

    Pitman, Jim and Yor, Marc , title =. The Annals of Probability , volume =

  76. [76]

    Scientific Reports , volume =

    Aletti, Giacomo and Crimaldi, Irene , title =. Scientific Reports , volume =

  77. [77]

    Scientific Reports , volume=

    Interacting innovation processes , author=. Scientific Reports , volume=. 2023 , publisher=

    2023

  78. [78]

    arXiv preprint arXiv:2501.09648 , year=

    Central limit theorems for interacting innovation processes, related statistical tools and general results , author=. arXiv preprint arXiv:2501.09648 , year=

    work page arXiv

  79. [79]

    Physical Review Research , volume=

    Full spectrum of novelties in time-dependent urn models , author=. Physical Review Research , volume=. 2025 , publisher=

    2025

  80. [80]

    The origin of bursts and heavy tails in human dynamics , journal =

    Barab. The origin of bursts and heavy tails in human dynamics , journal =

Showing first 80 references.