Pith. sign in

REVIEW 2 major objections 6 minor 54 references

DIT: Dimension Reduction View on Optimal NFT Rarity Meters

T0 review · 2 major / 6 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read A rarity meter fit to trade data beats five NFT rivals.

desk verdict The paper's headline superiority claim is compromised by evaluating DIT under the very objective it minimizes, but the MDS framing and public code are real assets. read the letter →

arxiv 2508.12671 v1 pith:PWXGEPD2 submitted 2025-08-18 cs.DC cs.LG

Dmitry Belousov , Yury Yanovich This is my paper · ORCID
classification cs.DCcs.LG
keywords NFTraritymetersnon-metricweightedunidimensionalscalingdimensionreductionDissimilarityinTrades(DIT)ROARbenchmarkstressblockchain
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 claims that NFT rarity is best modeled as a one-dimensional summary of pairwise trade dissimilarities, and that this point of view yields both a better meter and a cheaper way to evaluate meters. It introduces Dissimilarity in Trades (DIT), a rarity score obtained by minimizing the non-metric weighted unidimensional scaling stress $F$ between pairwise rarity distances and dissimilarities built from close-in-time NFT deals. On the ROAR benchmark of 100 collections, DIT achieves the best $F$ on over 65% of collections, ahead of Rarity.tools, KRAMER, OpenRarity, NFTGo, and ROAR. DIT's scores are non-interpretable: they are optimized numbers rather than transparent trait formulas. The paper's new performance measure $F$ works on $N\times N$ matrices after one aggregation of trade pairs, which it argues makes large-scale evaluation much faster than the earlier weighted-correlation measure.

What carries the argument

The load-bearing object is the NWUDS stress $F$ with its dissimilarity matrix $\delta_{ij}$ and weight matrix $w_{ij}$, where each weight aggregates a time-kernel across close-in-time deals of the same token pair and each dissimilarity is the weighted mean absolute log-price gap for that pair. The optimization follows Pliner's smoothing construction: $S_\varepsilon(\mathbf{x})$ is the raw stress averaged over an $\varepsilon$-cube, and the minimum is approached by the fixed-point iteration $x^{q+1}_i = \sum_j w_{ij}(x^q_j - d_{ij}u_\varepsilon(x^q_i-x^q_j))/\sum_j w_{ij}$, with $\varepsilon$ annealed from $2d^*$ toward zero and a final local minimization of the raw stress. The proof of convergence identifies the iteration with gradient descent at step size $1/(2N)$ and invokes Polyak's theorem; Theorem 2 gives a regime ($\varepsilon > 4d^*$) where the smoothed stress has a unique minimum. Out-of-sample tokens receive DIT scores by kernel regression on an interpretable rarity meter $\tilde{R}$, with the neighbor count $k$ chosen by cross-validation.

What would settle it

Rerun the ROAR comparison with the established weighted-correlation measure $F_{wc}$ on the same held-out trades and scale-adjusted meters; if DIT no longer wins on most collections, its reported edge comes from optimizing the benchmark rather than from market alignment. A second check is to take labeled wash-traded or manipulated NFTs and see whether DIT's most extreme scores flag them more accurately than the interpretable meters.

Watch

Extended reading notes

Core claim

The paper's central claim is that the rarity-meter design problem reduces to non-metric weighted unidimensional scaling (NWUDS): from trade data one builds weighted dissimilarities $\delta_{ij}$ between token pairs, and an optimal rarity vector $\mathbf{R}$ minimizes the normalized stress $$F(\mathbf{R}) = \sqrt{ \frac{\sum_{i,j} w_{ij}(|R_i-R_j|-\delta_{ij})^2}{\sum_{i,j} w_{ij}|R_i-R_j|^2} }.$$ It argues that this objective is the right market-alignment criterion, because rarity gaps should reproduce price gaps between deals that happen close in time, and that smoothing the stress over a cube of side $\varepsilon$ turns the multiextremal problem into one that annealing can solve. The resulting DIT meter is non-interpretable by design, yet on the test portion of the ROAR benchmark it records the best $F$ on more than 65% of the 100 collections, with ROAR and NFTGo following. A supporting contribution is the claim that $F$ itself is a better practical performance measure than the established weighted correlation $F_{wc}$, because it is cheaper to compute.

Load-bearing premise

The load-bearing premise is that the gap between two tokens' rarity scores ought to reproduce the gap between their close-in-time trade prices; the paper then scores every competitor with that same yardstick, which is also the yardstick its own meter was built to minimize.

Editorial extensions

If this is right

  • If $F$ is accepted as the criterion, DIT's scores track the pairwise price structure of NFT trades more closely than any of the five interpretable meters tested.
  • Because $F$ needs only the aggregated $N\times N$ dissimilarity and weight matrices, evaluating a rarity meter on large collections is much cheaper than recomputing over pairwise deals, which makes benchmarking on big collections practical.
  • DIT's loss of interpretability is not fatal for automation: the scores can be fed into ranking, pricing, and anomaly-detection tools, and the paper argues they are suitable for spotting wash trading and pump-and-dump patterns.
  • The same construction carries over to any one-dimensional score meant to mirror pairwise dissimilarities, so the framework is reusable beyond NFT collections.

Reading between the lines

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

  • The benchmark is tilted toward DIT: all competitors are scored with $F$, the same objective DIT was optimized to minimize, so the claimed margin may shrink or vanish under the earlier weighted-correlation measure $F_{wc}$ or an independently defined price target.
  • Because $k=2$ is the optimal neighbor count in roughly 60% of collections, a fixed default of two neighbors might be nearly as good as per-collection selection, but the paper does not report that comparison.
  • The paper's wash-trading and market-manipulation applications are forward-looking rather than tested; a direct test would compare DIT's outlier scores against labeled instances of known manipulation.
  • The dimension-reduction recipe should transfer to non-blockchain collectibles or any asset with transaction histories, though the paper only demonstrates it on NFTs.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

2 major / 6 minor

Summary. The paper proposes a dimension-reduction formulation of NFT rarity. Given trade-derived pairwise dissimilarities δ_ij with weights w_ij, it defines a normalized weighted stress F(R) between the rarity-induced distances d_ij(R)=|R_i-R_j| and δ_ij. The main methodological contribution is DIT, a rarity meter obtained by minimizing F through a smoothed iterative algorithm for non-metric weighted unidimensional scaling, together with a scale-adjustment formula (Statement 1) and an out-of-sample kernel extension. Using the ROAR benchmark of 100 collections, the authors compare DIT with Rarity.tools, KRAMER, OpenRarity, NFTGo, and ROAR and report that DIT achieves the best F-based performance in over 65% of collections. The paper claims that DIT is 'optimal' in the NWUDS sense and that it 'demonstrates superior performance compared to existing methods.'

Significance. If the reported superiority could be established under an externally validated performance measure, the paper would make a useful contribution: it supplies a clean mathematical formulation, a correct scale-adjustment derivation in Statement 1, a reproducible open-source benchmark, and a computationally cheaper objective than the weighted correlation used by ROAR. The dimension-reduction link between rarity and trade dissimilarities is a reasonable research direction with potential applications beyond NFTs. At present, however, the headline empirical claim is not independently established: it rests on a metric that the proposed method itself optimizes, and no evidence is given that this metric ranks meters similarly to the established weighted correlation Fwc or predicts market outcomes.

major comments (2)
  1. [Section 7.2 and Figure 2] The head-to-head comparison is circular with respect to the proposed objective. DIT is trained by minimizing F (Section 5, Algorithm 1), and Section 7.2 then adopts 'the NWUDS-inspired DIT's F as the performance measure, replacing the previously employed weighted correlation Fwc.' The competitors are only rescaled through Statement 1; they are not optimized for F. In these circumstances, DIT winning under F is expected and does not constitute independent evidence of market alignment. The manuscript provides no validation that F orders meters similarly to Fwc or any other external criterion. Please repeat the performance-profile analysis under Fwc, report the Fwc values for all six meters, and additionally report correlations between F and Fwc across collections; without this, the 'superior performance' claim in the abstract and Section 7.3 is not a falsifiable finding about NFT rarity.
  2. [Section 5, Theorems 1-3 and Section 8] The paper overstates its optimality guarantee. Theorem 1 establishes monotone decrease and convergence to a stationary point of the smoothed stress S_ε; Theorem 2 gives a condition under which the smoothed problem has a unique stationary point at zero; Theorem 3 gives an ε→0 approximation result. None of these implies that the final local minimization in Algorithm 1 reaches a global minimum of the original stress S. The sentence in Section 5 that the procedure 'ensures that the solution converges to a global minimum' and the Section 8 statement about 'convergence ... to a globally optimal solution' are therefore unsupported. Please either remove the global-optimality claims or provide a rigorous guarantee.
minor comments (6)
  1. [Section 3.2.2] In the KRAMER definition, the weights are denoted α_1,...,α_N, but the sum runs over T traits; this should be α_1,...,α_T.
  2. [Section 5] In the suggested ε-sequence, the formula for d* uses the index I in both the maximum and the sums; the collection size is N, so N should replace I.
  3. [Section 7.3] The name of the fourth comparator appears as 'NFTGO' in the text and 'NFTGo' in Figure 2; please harmonize the spelling.
  4. [Section 7.4] The text says the optimal k values ranged from 1 to 25, but Figure 3 displays only k=2,...,10; either expand the figure or clarify the range shown.
  5. [Section 7.4] The statement 'the optimal k was 2 in 60%' should specify that this is the fraction of collections, not of tokens or trades.
  6. [Section 6.2 and Section 7.2] The out-of-sample description would benefit from an explicit statement that the dissimilarity and weight matrices used in Algorithm 1 are computed from the training portion of the trades only, and that Figure 2's F values are computed on the test portion.

Circularity Check

1 steps flagged · score 6.0 of 10

The benchmark measure F is the same objective DIT is constructed to minimize; Section 7.2 replaces ROAR's Fwc with F, so the claimed superiority is largely a statement about DIT's own loss rather than independent market alignment.

  1. self definitional [Section 7.2, Performance Measure; Sections 4-5 define F and DIT]
    "We adopt the NWUDS-inspired DIT's F as the performance measure, replacing the previously employed weighted correlation Fwc."

    Section 4 defines F as the normalized NWUDS stress between pairwise rarity distances and trade dissimilarities, and Section 5 states that the optimal rarity meter under F minimizes this stress. The DIT algorithm is explicitly constructed as that minimizer. Section 7.2 then makes F the official benchmark for all meters and Figure 2 reports performance profiles under F, after rescaling competitors with Statement 1's F-minimizing alpha. Therefore the headline result that DIT outperforms competitors restates, up to out-of-sample smoothing, that DIT was optimized for the same criterion on which it is judged. Because F is never validated against the established weighted correlation Fwc or against external outcomes, the evaluation does not provide independent evidence of market alignment.

full rationale

The central issue is not that DIT minimizes an objective—that is standard optimization—but that the paper then adopts that same objective as the evaluation benchmark, replacing the established ROAR measure Fwc without independent validation. The only mitigating factor is the train/test split and out-of-sample smoothing in Section 6.2: F is computed on held-out test trades, so the comparison is not statistically forced in the strictest sense. Nevertheless, the choice of metric completely determines the comparison, and since DIT is by construction the minimizer of F while the competitors are not, the superiority claim in Section 7.3 is substantially self-referential. This is a partial circularity rather than a fully forced result, because the empirical magnitude and the held-out generalization still contain information. The lack of external validation of F is a correctness/external-validity concern beyond circularity, but it amplifies the problem: without it, 'DIT is optimal' and 'DIT wins the benchmark' are the same statement.

Assumptions & free parameters 5 free parameters · 5 assumptions · 0 invented entities

The paper introduces no new physical or conceptual entities; it reuses trade data and existing rarity meters. The free parameters above are the quantities that must be fitted or chosen by hand for the central result to be reproduced.

free parameters (5)
  • alpha scale factor = per meter per collection, from Statement 1
    Each rarity meter is rescaled by alpha = sum(w delta^2) / sum(w d delta) to minimize F, so every comparison depends on this fitted scalar.
  • k (neighborhood size) = 1 to 25, optimal k=2 in 60% of collections
    Selected per collection by grid search during cross-validation for the out-of-sample extension (Section 6.2).
  • base interpretable meter R-tilde = NFTGo in most collections
    Selected per collection by grid search; affects the smoothed rarity scores for new tokens.
  • initial configuration x0 = unspecified
    The continuation algorithm's outcome depends on the initial configuration, which is not fixed in the paper.
  • smoothing schedule parameters (epsilon sequence, Q) = unspecified
    Algorithm 1 depends on epsilon_1, Q, and the local minimization routine, none of which are specified for the experiments.
assumptions (5)
  • standard math Polyak's convergence theorem for gradient descent with constant step size
    Used in the proof of Theorem 1 to claim convergence to a stationary point.
  • domain assumption The weighted absolute log-price ratio of close-in-time deals is a valid dissimilarity measure for NFT rarity
    Introduced in Section 4; the entire F measure is built on this assumption, with no external validation.
  • domain assumption F is a valid performance measure for rarity meters
    Section 7.2 replaces the established Fwc with F without showing they agree; all later conclusions depend on this.
  • domain assumption The trait space of NFTs is smooth enough for kernel-based out-of-sample extension
    Section 6.2 applies nonparametric regression on |R_tilde(Xn)-R_tilde(X)|, implicitly assuming latent manifold structure.
  • domain assumption Pairs of trades close in time are directly comparable despite market volatility
    Section 4 uses the kernel k(t1,t2) to weight close deals, assuming volatility is mitigated by temporal proximity.

how reviews work

0 comments
Cite this review

Pith. "Pith review of DIT: Dimension Reduction View on Optimal NFT Rarity Meters." pith.science (2026). https://pith.science/paper/PWXGEPD2

@misc{pith2026250812671,
  author       = {Pith},
  title        = {Pith review of: DIT: Dimension Reduction View on Optimal NFT Rarity Meters},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/PWXGEPD2}},
  note         = {Machine review of arXiv:2508.12671}
}
read the original abstract

Non-fungible tokens (NFTs) have become a significant digital asset class, each uniquely representing virtual entities such as artworks. These tokens are stored in collections within smart contracts and are actively traded across platforms on Ethereum, Bitcoin, and Solana blockchains. The value of NFTs is closely tied to their distinctive characteristics that define rarity, leading to a growing interest in quantifying rarity within both industry and academia. While there are existing rarity meters for assessing NFT rarity, comparing them can be challenging without direct access to the underlying collection data. The Rating over all Rarities (ROAR) benchmark addresses this challenge by providing a standardized framework for evaluating NFT rarity. This paper explores a dimension reduction approach to rarity design, introducing new performance measures and meters, and evaluates them using the ROAR benchmark. Our contributions to the rarity meter design issue include developing an optimal rarity meter design using non-metric weighted multidimensional scaling, introducing Dissimilarity in Trades (DIT) as a performance measure inspired by dimension reduction techniques, and unveiling the non-interpretable rarity meter DIT, which demonstrates superior performance compared to existing methods.

Figures

Figures reproduced from arXiv: 2508.12671 by the authors.

Figure 1
Figure 1. NFT n = 173 is part of the Bored Ape Yacht Club (BAYC) collection, which consists of 10, 000 unique tokens. The contract address for this collection is 0xbc4ca0eda7647a8ab7c2061c2e118a18a936f13d. BAYC tokens have seven distinct traits: background, clothes, eyes, hats, earrings, fur, and mouth. For NFT n = 173, the trait values are as follows: Yellow, Guayabera, Bored, Bayc Flipped Brim, None, Blue, and Bored. The re… view at source ↗
Figure 2
Figure 2. Performance profiles of rarity meters on the updated [PITH_FULL_IMAGE:figures/full_fig_p008_2.png] view at source ↗
Figure 3
Figure 3. Impact of the neighborhood parameter k on the perfor￾mance of DIT. 7.4. Impact of Neighborhood Parameter We investigate the effect of the neighborhood parameter k on the performance of DIT. This parameter controls the number of neighbors used in the nonparametric regression for out-of-sample extension [PITH_FULL_IMAGE:figures/full_fig_p009_3.png] view at source ↗

Review history (2 revisions) →

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

54 extracted references · 37 canonical work pages

  1. [1]

    Oliveira, I

    L. Oliveira, I. Bauer, L. Zavolokina, G. Schwabe, To token or not to token: Tools for understanding blockchain tokens, in: International Conference on Information Systems 2018, ICIS 2018, 2018, pp. 1–17

    work page 2018

  2. [2]

    di Angelo, G

    M. di Angelo, G. Salzer, Tokens, Types, and Standards: Identifi- cation and Utilization in Ethereum, in: 2020 IEEE International Conference on Decentralized Applications and Infrastructures (DAPPS), IEEE, 2020, pp. 1–10. doi:10.1109/DAPPS49028. 2020.00001. URL https://ieeexplore.ieee.org/document/9126009/

    arXiv 2020

  3. [3]

    N. Wang, S. Chi-Kin Chau, Y. Zhou, Privacy-Preserving En- ergy Storage Sharing with Blockchain, in: Proceedings of the Twelfth ACM International Conference on Future Energy Sys- tems, ACM, New York, NY, USA, 2021. URL https://doi.org/10.1145/3447555.3464869

    arXiv 2021

  4. [4]

    Wood, Polkadot: Vision for a Heterogeneous Multi-Chain Framework, Whitepaper (2017) 1–21 doi: 10.1021/acs.jpcc.6b00269

    G. Wood, Polkadot: Vision for a Heterogeneous Multi-Chain Framework, Whitepaper (2017) 1–21 doi: 10.1021/acs.jpcc.6b00269. URL https://github.com/w3f/polkadot-white-paper/raw/ master/PolkaDotPaper.pdf

    work page doi:10.1021/acs.jpcc.6b00269 2017

  5. [5]

    Nakamoto, Bitcoin: A Peer-to-Peer Electronic Cash System, www.bitcoin.org (2008) 1–9

    S. Nakamoto, Bitcoin: A Peer-to-Peer Electronic Cash System, www.bitcoin.org (2008) 1–9. URL https://bitcoin.org/bitcoin.pdf

    work page 2008

  6. [6]

    Yakovenko, Solana: A new architecture for a high perfor- mance blockchain (2018)

    A. Yakovenko, Solana: A new architecture for a high perfor- mance blockchain (2018). URL https://solana.com/solana-whitepaper.pdf

    work page 2018

  7. [7]

    URL https://raritytools.medium.com/ranking-rarity- understanding-rarity-calculation-methods-86ceaeb9b98c

    Rarity.tools, Ranking Rarity: Understanding Rarity Calcula- tion Methods (2021). URL https://raritytools.medium.com/ranking-rarity- understanding-rarity-calculation-methods-86ceaeb9b98c

    work page 2021

  8. [8]

    Krasnoselskii, Y

    M. Krasnoselskii, Y. Madhwal, Y. Yanovich, KRAMER: Inter- pretable Rarity Meter for Crypto Collectibles, IEEE Access 11 (2023) 4283–4290. doi:10.1109/ACCESS.2023.3236080. URL https://ieeexplore.ieee.org/document/10014994/

    arXiv 2023

Show all 54 references
  1. [9]

    Belousov, M

    D. Belousov, M. Shuklin, A. Stepin, Y. Yanovich, ROAR: A Benchmark for NFT Rarity Meters, in: 2024 IEEE International Conference on Blockchain and Cryptocurrency (ICBC), IEEE, 2024

    2024

  2. [10]

    Paule-Vianez, A

    J. Paule-Vianez, A. Alc´ azar-Blanco, J. L. Coca-P´ erez, Effect of Economic Policy Uncertainty on the investment in numismatic assets: Evidence for the Walking Liberty Half Dollar, Finance Research Letters 46 (2022) 102412. doi:10.1016/J.FRL.2021. 102412

    2022 doi

  3. [11]

    Chailan, Art as a means to recreate luxury brands’ rarity and value, Journal of Business Research 85 (2018) 414–423

    C. Chailan, Art as a means to recreate luxury brands’ rarity and value, Journal of Business Research 85 (2018) 414–423. doi: 10.1016/J.JBUSRES.2017.10.019

    2018 doi

  4. [12]

    Koford, A

    K. Koford, A. E. Tschoegl, The market value of rarity, Journal of Economic Behavior & Organization 34 (3) (1998) 445–457. doi:10.1016/S0167-2681(97)00084-X

    1998 doi

  5. [13]

    Hastie, R

    T. Hastie, R. Tibshirani, J. Friedman, The Elements of Statis- tical Learning, Springer Series in Statistics, Springer New York, New York, NY, 2009. doi:10.1007/978-0-387-84858-7

    2009 doi

  6. [14]

    J. E. Hughes, Demand for Rarity: Evidence from a Collectible Good, The Journal of Industrial Economics 70 (1) (2022) 147–

    2022

  7. [15]

    URL https://sourceforge.net/software/product/rarity

    Slashdot Media, Best rarity.tools Alternatives & Competitors (2022). URL https://sourceforge.net/software/product/rarity. tools/alternatives

    2022

  8. [16]

    Krasnoselskii, Y

    M. Krasnoselskii, Y. Madhwal, Y. Yanovich, KRAMER: Ka- naria NFT Collection Rarity Meter, in: 2022 IEEE Interna- tional Conference on Blockchain and Cryptocurrency (ICBC), IEEE, 2022, pp. 1–2. doi:10.1109/ICBC54727.2022.9805542. URL https://ieeexplore.ieee.org/document/9805542/

    2022

  9. [17]

    White, A

    B. White, A. Mahanti, K. Passi, Characterizing the OpenSea NFT Marketplace, in: Companion Proceedings of the Web Con- ference 2022, ACM, New York, NY, USA, 2022, pp. 488–496. doi:10.1145/3487553.3524629

    2022

  10. [18]

    URL https://mirror.xyz/openrarity.eth/- R8ZA5KCMgqtsueySlruAhB77YBX6fSnS_dT-8clZPQ

    OpenRarity, Introducing OpenRarity (2022). URL https://mirror.xyz/openrarity.eth/- R8ZA5KCMgqtsueySlruAhB77YBX6fSnS_dT-8clZPQ

    2022

  11. [19]

    F. Yan, X. Wang, K. Mao, W. Zhang, W. Chen, NFTVis: Vi- sual Analysis of NFT Performance, in: 2023 IEEE 16th Pacific Visualization Symposium (PacificVis), IEEE, 2023, pp. 82–91. doi:10.1109/PacificVis56936.2023.00016

    2023

  12. [20]

    URL https://docs.nftgo.io/

    NFTGo, NFTGo: The World’s Leading NFT Data Intelligence Provider. URL https://docs.nftgo.io/

  13. [21]

    Lee, G.-C

    H. Lee, G.-C. Lee, H.-Y. Koo, Exploring the relationship be- tween rarity and price of profile picture NFT: A formal concept analysis on the BAYC NFT collection, Blockchain: Research and Applications (2024). doi:10.1016/j.bcra.2024.100191

    2024

  14. [22]

    Nadini, L

    M. Nadini, L. Alessandretti, F. Di Giacinto, M. Martino, L. M. Aiello, A. Baronchelli, Mapping the NFT revolution: market trends, trade networks, and visual features, Scientific Reports 11 (1) (2021) 20902. doi:10.1038/s41598-021-00053-8

    2021 doi

  15. [23]

    Mekacher, A

    A. Mekacher, A. Bracci, M. Nadini, M. Martino, L. Alessan- dretti, L. M. Aiello, A. Baronchelli, Heterogeneous rarity pat- terns drive price dynamics in NFT collections, Scientific Reports 12 (1) (2022) 13890. doi:10.1038/s41598-022-17922-5 . URL https://www.nature.com/article...

    2022 doi

  16. [24]

    Schaar, S

    L. Schaar, S. Kampakis, Non-fungible Tokens as an Alter- native Investment: Evidence from CryptoPunks, The Jour- nal of The British Blockchain Association 5 (1) (2022) 1–12. doi:10.31585/jbba-5-1-(2)2022

    2022 doi

  17. [25]

    Aydo˘ gdu, N

    D. Aydo˘ gdu, N. Aydin, Development of a hybrid recommen- dation system for nfts using deep learning techniques, IEEE Access (2024). doi:10.1109/ACCESS.2024.3514512. URL https://ieeexplore.ieee.org/document/10786976/

    2024

  18. [26]

    M. Pala, E. Sefer, Nft price and sales characteristics prediction by transfer learning of visual attributes, The Journal of Finance and Data Science 10 (2024) 100148. doi:10.1016/j.jfds.2024. 100148

    2024 doi

  19. [27]

    Pearson, On lines and planes of closest fit to systems of points in space, Philosophical Magazine 2 (11) (1901) 559–572

    K. Pearson, On lines and planes of closest fit to systems of points in space, Philosophical Magazine 2 (11) (1901) 559–572. doi:10.1080/14786440109462720

    1901 doi

  20. [28]

    M. A. A. Cox, T. F. Cox, Multidimensional Scaling, in: Handbook of Data Visualization, Springer Berlin Heidelberg, Berlin, Heidelberg, 2008, pp. 315–347. doi:10.1007/978-3- 540-33037-0{\_}14. URL http://link.springer.com/10.1007/978-3-540-33037- 0_14

    2008 doi

  21. [29]

    J. d. Leeuw, P. Mair, Multidimensional Scaling Using Majoriza- tion: SMACOF in R, Journal of Statistical Software 31 (3) (2009). doi:10.18637/jss.v031.i03

    2009 doi

  22. [30]

    H. S. Seung, The Manifold Ways of Perception, Science 290 (5500) (2000) 2268–2269. doi:10.1126/science.290.5500. 2268. URL http://www.sciencemag.org/cgi/doi/10.1126/science. 290.5500.2268

    2000

  23. [31]

    Levina, P

    E. Levina, P. J. Bickel, Maximum Likelihood Estimation of In- trinsic Dimension, in: Advances in Neural Information Process- ing Systems, MIT Press, 2005, pp. 777–784. URL https://www.stat.berkeley.edu/~bickel/mldim.pdf

    2005

  24. [32]

    Gomtsyan, N

    M. Gomtsyan, N. Mokrov, M. Panov, Y. Yanovich, Geometry- Aware Maximum Likelihood Estimation of Intrinsic Dimension, Proceedings of Machine Learning Research 101 (2019) 1126– 1141. URL http://arxiv.org/abs/1904.06151

    2019 arXiv

  25. [33]

    Belkin, P

    M. Belkin, P. Niyogi, Laplacian Eigenmaps for Dimensionality Reduction and Data Representation, Neural Computation 10 15 (6) (2003) 1373–1396. doi:10.1162/089976603321780317. URL https://direct.mit.edu/neco/article/15/6/1373- 1396/6730

    2003 doi

  26. [34]

    X. Huo, X. S. Ni, A. K. Smith, A Survey of Manifold- Based Learning Methods, in: Recent Advances in Data Mining of Enterprise Data: Algorithms and Ap- plications, WORLD SCIENTIFIC, 2008, pp. 691–745. doi:10.1142/9789812779861{\_}0015. URL http://www.worldscientific.com/doi/abs...

    2008 doi

  27. [35]

    Singer, H.-T

    A. Singer, H.-T. Wu, Vector diffusion maps and the connection Laplacian, Communications on Pure and Applied Mathematics 65 (8) (2012) 1067–1144. doi:10.1002/cpa.21395. URL http://doi.wiley.com/10.1002/cpa.21395

    2012 doi

  28. [36]

    Bernstein, A

    A. Bernstein, A. Kuleshov, Y. Yanovich, Information preserv- ing and locally isometric&conformal embedding via Tan- gent Manifold Learning, in: Proceedings of the 2015 IEEE In- ternational Conference on Data Science and Advanced Analyt- ics (DSAA), IEEE, 2015, pp. 1–9. doi...

    2015

  29. [37]

    Baldi, Autoencoders, Unsupervised Learning, and Deep Ar- chitectures, ICML Unsupervised and Transfer Learning (2012)

    P. Baldi, Autoencoders, Unsupervised Learning, and Deep Ar- chitectures, ICML Unsupervised and Transfer Learning (2012). doi:10.1561/2200000006

    2012 doi

  30. [38]

    D. Bank, N. Koenigstein, R. Giryes, Autoencoders, Springer International Publishing, 2023, pp. 353–374. doi:10.1007/978- 3-031-24628-9_16

    2023 doi

  31. [39]

    B. L. Ross, G. Loaiza-Ganem, A. L. Caterini, J. C. Cresswell, Neural Implicit Manifold Learning for Topology-Aware Density Estimation (6 2022)

    2022

  32. [40]

    V. M. Pliner, A class of metric scaling models, Avtomat. i Tele- mekh. (1984) 122–128

    1984

  33. [41]

    V. M. Pliner, The problem of multidimensional metric scaling, Avtomat. i Telemekh. (1986) 140–148

    1986

  34. [42]

    Pliner, Metric unidimensional scaling and global optimiza- tion, J

    V. Pliner, Metric unidimensional scaling and global optimiza- tion, J. Classification 13 (1) (1996) 3–18. doi:10.1007/ BF01202579

    1996

  35. [43]

    Y. Ge, Y. Leung, J. Ma, Unidimensional scaling classifier and its application to remotely sensed data, in: Proceedings. 2005 IEEE International Geoscience and Remote Sensing Sympo- sium, 2005. IGARSS ’05., IEEE, 2005, pp. 3841–3844. doi: 10.1109/IGARSS.2005.1525747

    2005

  36. [44]

    L. J. Hubert, P. Arabie, J. J. Meulman, Linear Unidimensional Scaling in the L 2 -Norm: Basic Optimization Methods Using MATLAB, Journal of Classification 19 (2) (2002) 303–328. doi: 10.1007/s00357-001-0047-5

    2002 doi

  37. [45]

    Guttman, A general nonmetric technique for finding the smallest coordinate space for a configuration of points, Psy- chometrika 33 (4) (1968) 469–506

    L. Guttman, A general nonmetric technique for finding the smallest coordinate space for a configuration of points, Psy- chometrika 33 (4) (1968) 469–506. doi:10.1007/BF02290164

    1968 doi

  38. [46]

    M. J. Brusco, A Simulated Annealing Heuristic for Unidimen- sional and Multidimensional (City-Block) Scaling of Symmet- ric Proximity Matrices, J. Classification 18 (1) (2001) 3–33. doi:10.1007/s00357-0003-4

    2001 doi

  39. [47]

    Polyak, Introduction to Optimization, Nauka, 1983

    B. Polyak, Introduction to Optimization, Nauka, 1983

    1983

  40. [48]

    C. M. Bishop, Pattern Recognition and Machine Learning, Springer-Verlag, New York, 2006

    2006

  41. [49]

    Bengio, J.-F

    Y. Bengio, J.-F. Paiement, P. Vincent, Out-of-Sample Exten- sions for LLE, Isomap, MDS, Eigenmaps, and Spectral Clus- tering, In Advances in Neural Information Processing Systems (2003) 177–184doi:10.1.1.5.1709

    2003

  42. [50]

    Yanovich, Asymptotic Properties of Nonparametric Estima- tion on Manifold, JMLR Workshop and Conference Proceedings 60 (2017) 18–38

    Y. Yanovich, Asymptotic Properties of Nonparametric Estima- tion on Manifold, JMLR Workshop and Conference Proceedings 60 (2017) 18–38

    2017

  43. [51]

    URL https://github.com/fasghq/DIT-benchmark/tree/main

    fasghq, maximshuklin, Dit-benchmark: A benchmark for nft rarity meters (2025). URL https://github.com/fasghq/DIT-benchmark/tree/main

    2025

  44. [52]

    M. L. Morgia, A. Mei, A. M. Mongardini, E. N. Nemmi, A Game of NFTs: Characterizing NFT Wash Trading in the Ethereum Blockchain, in: 2023 IEEE 43rd International Conference on Distributed Computing Systems (ICDCS), IEEE, 2023, pp. 13–

    2023

  45. [54]

    doi:10.1109/ICDCS57875.2023.00018. 11

    2023

  46. [167]

    URL https://onlinelibrary.wiley.com/doi/10.1111/joie

    doi:10.1111/JOIE.12262. URL https://onlinelibrary.wiley.com/doi/10.1111/joie. 12262

    doi

Pith tools

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