Pith. sign in

REVIEW 3 major objections 7 minor 50 references

Rate-Distortion-Perception Theory: Redefining the Fundamental Limits of Information Representation

T0 review · 3 major / 7 minor · reviewed 2026-08-01 · deepseek-v4-flash

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

Pith's one-line read The rate-distortion-perception function can be computed, not just defined from first principles.

desk verdict A competent tutorial that repackages the authors' own results; the title oversells it and the RAM convergence condition is not constructively instantiated, but as a survey it deserves referee time. read the letter →

arxiv 2607.17232 v1 pith:IJDQHMGX submitted 2026-07-19 cs.IT cs.LGcs.SYeess.SYmath.IT

Photios A. Stavrou , Giuseppe Serra , Marios Kountouris This is my paper
classification cs.ITcs.LGcs.SYeess.SYmath.IT MSC 94A3494A1762B10
keywords rate-distortion-perceptionfunctionlossycompressionperceptualfidelityf-divergencealternatingminimizationGaussiansourcesperfectrealismcopulamethods
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 tutorial argues that the rate-distortion-perception function (RDPF), which gives the minimum bit rate for a lossy code whose reconstructions are both close to the source and statistically similar to it, is a computable object. For discrete sources with f-divergence perception constraints, it assembles alternating-minimization algorithms, including a Newton-based variant and a relaxed variant, and claims their iterates converge to the RDPF. For Gaussian sources, it gives closed-form expressions under alpha-divergence and squared optimal-transport perception, plus a water-filling interpretation. For continuous non-Gaussian sources under perfect realism, it develops a copula-based convex program that can be solved numerically. The reader should care because these algorithms turn a three-way trade-off central to modern generative compression into concrete numerical curves that can be measured, benchmarked, and designed against.

What carries the argument

The workhorse is the Lagrangian double-minimization reformulation of the RDPF: minimizing mutual information is replaced by minimizing a KL divergence plus distortion and perception penalty terms over a conditional distribution Q and an auxiliary reconstruction distribution h. The optimal h for fixed Q is the induced output distribution; the optimal Q for fixed h is a Boltzmann-like kernel weighted by an exponent involving the distortion and a function of the divergence. Because the h-update is implicit, the paper's two algorithms resolve it either by Jacobian-based root-finding (the NAM scheme) or by substituting an auxiliary functional (the RAM scheme). For Gaussian sources, the same machi

What would settle it

Run the relaxed alternating-minimization scheme on a Bernoulli source with Hamming distortion and total-variation perception, using the paper's own smooth approximation as a benchmark. If the relaxed scheme's limit misses the analytic RDPF in a regime where the linear-rate condition provably holds, the central convergence claim would be settled. A second observation: if a divergence satisfies the linear-rate condition but the scheme converges to different limits from different initial distributions, the claimed global optimality fails.

Watch

Extended reading notes

Core claim

The central claim is that the RDPF, as defined in (1), can be solved by alternating minimization. For finite-alphabet sources with f-divergence perception, the paper shows that an optimal alternating-minimization update converges to a point on the RDPF, and since the update is implicit, it develops two implementable variants: a root-finding iteration that uses the Jacobian of the update map and requires twice-differentiable divergences, and a relaxed iteration that introduces an auxiliary functional to make the update explicit while preserving global convergence under a linear-rate condition. For Gaussian sources, the paper derives parametric closed-form solutions for alpha-divergence and sq

Load-bearing premise

The relaxed alternating-minimization convergence theorem requires an auxiliary functional that predicts the implicit next iterate closely enough to converge at a linear rate, and the paper gives no constructive recipe for such a functional; if no such choice exists for a given divergence, the scheme's global-convergence guarantee collapses.

Editorial extensions

If this is right

  • If the convergence theorems hold, the RDPF for discrete sources under any smooth f-divergence can be computed pointwise by the Newton-based scheme with exponential convergence, and by the relaxed scheme without smoothness requirements, for the multiplier range where the linear-rate condition holds.
  • For Gaussian sources, the closed-form RDPF expressions under alpha-divergence and squared optimal-transport perception imply that optimal encoders take linear realizations with closed-form parameters, connecting information-theoretic limits to lattice-based coding.
  • In the perfect-realism regime, the Gaussian water-filling allocation changes from a single global water level to component-dependent levels, so every source component must remain represented even at low rates.
  • The copula-based estimator provides a numerical route to the perfect-realism RDPF for non-Gaussian continuous sources, complementing the entropy-type lower bound that is tight only in the low-distortion regime.
  • The paper's comparison with barycenter and primal-dual methods suggests these approaches cover complementary regimes of the RDPF landscape rather than competing for the same problems.

Reading between the lines

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

  • If the RDPF becomes routinely computable, perception-aware codec design can shift from heuristics to targeting a computed rate-perception-distortion frontier; one could benchmark a generative codec by its distance to this surface.
  • The relaxed scheme's reliance on an auxiliary functional is a gap: a constructive rule for choosing that functional would make the algorithm fully general, and one could test whether running a few inner root-finding steps to define the auxiliary map yields the required linear-rate convergence.
  • The copula formulation suggests a natural estimator for the RDPF under non-tensorizable distortion measures, since the projection is expressed on the copula space rather than on the original signal space, potentially connecting to entropic optimal-transport algorithms in neural compression.
  • The Gaussian closed forms show that perfect-realism coding allocates rate to preserve marginal laws rather than only to reduce squared error, a design principle that could be tested in neural codecs by measuring per-component rate allocation against the water-filling prediction.
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

3 major / 7 minor

Summary. This tutorial surveys rate-distortion-perception (RDP) theory, centered on the RDPF of Blau and Michaeli as defined in Eq. (1). It reviews coding theorems and then presents optimization methods for computing the RDPF: Newton-based alternating minimization (NAM) and relaxed alternating minimization (RAM) for finite alphabets with f-divergence perception constraints, Gauss-Seidel/water-filling methods for Gaussian sources under alpha-divergence and squared Wasserstein-2 constraints, and a copula-based method for continuous sources in the perfect-realism regime. The paper also discusses a Shannon lower bound for the perfect-realism RDPF and outlines future research directions. The central claim is that Eq. (1) is computable by these general-purpose tools for discrete and continuous sources.

Significance. If the presented algorithms are correct, the survey would be a useful reference, consolidating computational RDPF results that are currently scattered across the authors' prior papers and related literature. The paper is candid about several limitations, including Remark 1 on NAM-versus-RAM, the TV subsection's admission that RAM does not fully recover the RDP surface, and Remark 5 on the scope of the Shannon lower bound. It also properly attributes the Gaussian Wasserstein closed form to [34] and compares with recent Wasserstein-barycenter and primal-dual approaches. However, the paper does not provide proofs or code, and some of its central computational claims—especially those involving RAM—are stated with hypotheses that are not instantiated. The tutorial's value therefore depends on how carefully these conditional results are delimited.

major comments (3)
  1. [Theorem 3 (RAM), Sec. III-A] The RAM result is the main mechanism offered for nondifferentiable perception metrics, but the convergence hypothesis is not operational. The theorem defines \hat Q, \hat h, and \upsilon^{(n)}=\upsilon[\hat h^{(n)}], yet the condition reads lim ||h^{(n+1)}-\upsilon^{(n)}||=0 with 'at least a linear rate'; h^{(n+1)} is undefined in the statement (only \hat h^{(n+1)} is defined) and \upsilon is allowed to be any distribution-to-distribution functional. Consequently the reader cannot check whether the hypothesis holds for the Bernoulli/TV examples in Figs. 2-3, and no constructive \upsilon is given for any divergence. Because the abstract and Section III present RAM as part of general-purpose tools for computing (1), this is a load-bearing gap. Please either instantiate \upsilon with a provable condition, or recast the RAM claims and Figs. 2-3 as conditional/heuristic.
  2. [Theorem 5, Eq. (19)] The claimed closed-form linear realization \hat X = aX + W is not explicit. In the first case a=\max\{1-D/\sigma_W^2,0\}, but \sigma_W^2 is defined two lines below in terms of a. This is a circular definition, so the 'closed-form design parameters' are not actually closed-form. The classical RD limit suggests the intended expression is a=\max\{1-D/\sigma_X^2,0\}; if so, please correct. As printed, the complement to [34] (the realizing test channel) is not verifiable.
  3. [Sec. III-B, Eqs. (20)-(22)] The reduction of the Gaussian vector RDPF to the separable program (22) asserts that the lower bound (21) 'holds with equality without loss of generality' via simultaneous diagonalization of (A,\Sigma_W,\Sigma_X). Simultaneous diagonalizability of the optimal solution is not shown; the text itself notes the inequality is strict beyond i.i.d. vectors. Since (22) underlies Theorem 6 and the Gauss-Seidel algorithm, either provide a proof or state precisely the condition under which equality is guaranteed, with a specific citation.
minor comments (7)
  1. [Definition 2] Convention (i): 'f(0)=lim_{x->0+} f(0)' should read 'f(0)=lim_{x->0+} f(x)'.
  2. [Lemma 1, Eq. (3)] The Lagrangian uses both h and q_{\hat X}; define q_{\hat X} before Eq. (3) and state that at the optimum h = q_{\hat X}, otherwise the notation is confusing.
  3. [Theorem 3] If the unadorned h in the convergence hypothesis is the OAM update from Theorem 1, say so explicitly and specify the norm; if it is \hat h, use consistent notation. Also define what 'at least a linear rate' means (linear in n, in which norm).
  4. [Table II] The RAM row lists 'Convergence Rate: O(log(1/epsilon))' and 'Total Complexity: O(log(1/epsilon))'. A rate is not a number of iterations; as printed this is dimensionally confused and omits dependence on the alphabet size and the cost of evaluating A[upsilon].
  5. [Sec. III-A, before Theorem 3] The sentence claiming that RAM 'still guarantees convergence to a globally optimal solution' overstates the conditional Theorem 3. It should be qualified with 'provided the hypothesis of Theorem 3 holds.'
  6. [Figs. 2-3] The captions do not state which RAM auxiliary functional v was used in the numerical studies. For a paper whose message is computability, this makes the figures hard to reproduce; please state the choice of v or describe the results as illustrative.
  7. [Sec. III-B, Fig. 8] The text refers to a 'bivariate exponential source with zero mean'. A genuine exponential marginal cannot have zero mean; specify the mean-shifted construction or change the wording.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the RDPF definition is external, the Gaussian Wasserstein closed form is externally attributed, and the computational theorems are conditional results rather than fitted predictions.

full rationale

The paper is a tutorial that compiles prior results. Definition 1 (the RDPF) is taken from Blau and Michaeli [2], an external source. The central computational claims in Section III are stated as theorems attributed to the authors' own prior work ([28], [32], [33], [39]), but they are not reductions to the target result. Theorem 1 states an alternating-minimization convergence result to R(D_s,P_s); Theorem 2 is a Newton root-finding result; Theorem 3 (RAM) is explicitly conditional on an auxiliary functional v satisfying lim ||h^{(n+1)} - v^{(n)}|| = 0 with linear rate. This condition is not instantiated (no constructive v is supplied), which is a real completeness/correctness limitation of the claimed general-purpose computability, but it is not circular: the condition does not define v in terms of R(D_s,P_s), and the theorem is a conditional stability result that would be valid for any v satisfying it. The squared-Wasserstein closed form in Theorem 5 is transparently attributed to [34, Thm. 1], with the authors adding only a linear test-channel realization obtained from KKT conditions. The copula-based results build on external results (Sklar [41], Csiszár [42]) and are sanity-checked against the Gaussian Shannon lower bound rather than fitted to the claimed output. Numerical figures (Figs. 2-3, 5-8) are illustrations or comparisons against known closed forms, not predictions of fitted parameters. The TV approximation is explicitly constructed with D_fn <= TV and D_fn -> TV, and the paper honestly notes in Remark 1 and the TV subsection that RAM may only recover part of the RDP curve. Overall, the derivation chain exhibits no step in which an output is equivalent by construction to an input, a fitted parameter is renamed a prediction, or a load-bearing conclusion rests solely on an unverified self-citation.

Assumptions & free parameters 1 free parameters · 6 assumptions · 0 invented entities

The tutorial contributes no new mathematics; all computational theorems are quoted from prior papers, half of them by the same authors. The central RDPF definition is external [2], and the Wasserstein Gaussian formula is attributed to [34], which reduces the circularity burden. But most of the originality load (NAM, RAM, Gauss-Seidel allocation, copula estimator, robust coding) rests on self-cited results that are not proven or implemented here.

free parameters (1)
  • Auxiliary functional υ[·] in RAM = None specified; 'any functional that maps a probability distribution to a probability distribution'
    Introduced in Theorem 3 as an approximation to the true update h(n+1). Convergence and global optimality of RAM depend on this hand-chosen degree of freedom, but no construction is given.
assumptions (6)
  • domain assumption Sources are i.i.d. with known distribution p_X and single-letter distortion/perception constraints
    Section III states the scope as 'i.i.d. sources under single-letter distortion constraints'; methods do not address memory or unknown source statistics.
  • standard math The divergence d(·||·) is convex in its second argument, so (1) is convex
    Invoked after Eq. (1) via [2]; needed for the double-minimization formulation in Lemma 1.
  • domain assumption The restated theorems from [28], [32], [33], and [39] are correct
    Theorems 1–3 and 4–7 and 10–13 are quoted without proof from these self-authored references; the tutorial inherits any gaps.
  • domain assumption For Gaussian problems, restricting to jointly Gaussian reconstructions / linear test channels is without loss of optimality
    Used in Eq. (20) and Theorems 4–7, relying on [32, Lemma 2]; exact closed forms depend on it.
  • ad hoc to paper There exists a RAM auxiliary functional υ[·] satisfying lim ||h(n+1)−υ(n)||=0 with at least linear rate
    Theorem 3 introduces this as a condition for convergence; no construction or verification is provided.
  • domain assumption The distortion and perception constraints are tensorizable in the multivariate Gaussian case
    Eq. (21) uses tensorizability via convex g,h; equality with pairwise commuting matrices is argued, but general validity is limited.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Rate-Distortion-Perception Theory: Redefining the Fundamental Limits of Information Representation." pith.science (2026). https://pith.science/paper/IJDQHMGX

@misc{pith2026260717232,
  author       = {Pith},
  title        = {Pith review of: Rate-Distortion-Perception Theory: Redefining the Fundamental Limits of Information Representation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/IJDQHMGX}},
  note         = {Machine review of arXiv:2607.17232}
}
read the original abstract

Classical rate-distortion (RD) theory has long established the fundamental limits of lossy compression by quantifying the minimum number of bits required to represent a source under a prescribed distortion constraint. However, widely used distortion measures such as mean-squared error often fail to capture perceptual quality or semantic validity, which are increasingly central in modern learning-driven applications. Rate-distortion-perception (RDP) theory extends the RD framework by introducing perception as a third fundamental axis, quantified via distributional similarity between the source and reconstructed signals, leading to the rate-distortion-perception function (RDPF). This tutorial provides a structured overview of the coding principles underlying perception-aware lossy compression and surveys recent achievability results under different randomness assumptions. It then presents a unifying optimization viewpoint for computing the RDPF as defined by Blau and Michaeli, for both discrete and continuous sources under broad families of perceptual constraints, including f-divergences, alpha-divergences, and Wasserstein-based metrics. Special attention is given to computational tools such as alternating minimization schemes, Newton-based methods, and convex optimization formulations, as well as to analytically tractable cases such as Gaussian sources and the perfect-realism regime. Unlike recent broad surveys that emphasize generative architectures and AI-empowered communication systems, this tutorial focuses on the coding-theoretic and computational machinery needed to characterize, compute, and interpret the RDP limits. Finally, the tutorial outlines promising research directions at the intersection of information theory, neural compression, robust source coding, and perception-aware networked control systems.

Figures

Figures reproduced from arXiv: 2607.17232 by the authors.

Figure 1
Figure 1. When applying rate–distortion–perception (RDP) theory to [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. R(D, P) for a Bernoulli source under Hamming distortion and Df (·||·) = DKL(·||·). provided that limn→∞ ||h (n+1) −υ (n) || = 0 with at least a linear rate of convergence. Theorem 3 facilitates the implementation of the alter￾nating minimization scheme by introducing an auxiliary variable υ[h (n) ], which serves as an approximation of the true update h (n+1) while remaining a function solely of the current iterate h… view at source ↗
Figure 3
Figure 3. RDPF under Hamming distortion and T V perception, computed with the RAM scheme, and under an approximation of TV via the sequence {Dfn (·||·) : n = 1, 2, . . .}, computed with the NAM scheme, for n ∈ {1, 10, 100}. To address this issue, we approximate the T V (·∥·) distance by a sequence of f-divergences {Dfn (·∥·) : n = 1, 2, . . .} such that Dfn (·∥·) → T V (·∥·) as n → ∞. Specifically, we consider a family of con… view at source ↗
Figures from the paper (7 more)
Figure 4
Figure 4. Figure 4: Polynomial (15) for α = −1.2, α = 0.1, α = 0.5, and α = 2, with the perception constraint fixed to P = 0.2. following result, which can be found in [32, Theorem 1], derives a parametric upper-bound expression for (1), hereinafter denoted by RG(D, P), under the assumpti…
Figure 5
Figure 5. Figure 5: R G(D, P) for a Gaussian source X ∼ N (0, ΣX) with ΣX = diag([1, 3, 5]) under a squared Wasserstein-2 distance. solve any problem of the form (20), provided that the perception constraint admits an analytical charac￾terization for jointly Gaussian random variables. As …
Figure 6
Figure 6. Figure 6: Comparison of the per-dimension distortion [PITH_FULL_IMAGE:figures/full_fig_p011_6.png]
Figure 7
Figure 7. Figure 7: b). We compare the results for various source distributions, such as Gaussian, Laplace, exponential, and uniform, assuming that the source X ∼ (0, 1), i.e., has zero mean and unit variance σ 2 X = 1. In Fig. 7a, we also compare the estimated result with the SLB derived…
Figure 8
Figure 8. Figure 8: RPR(D) under the MSE distortion metric for (a) a Gaussian, and (b) an exponential bivariate source. RPR(D) using [39, Algorithm 1] and compare it with the bound RSLB PR (D) in (42), where the term RG PR(D) is evaluated via the optimal adaptive reverse water-filling sol…
Figure 9
Figure 9. Figure 9: A distributionally robust scenario in image compression. [PITH_FULL_IMAGE:figures/full_fig_p017_9.png]
Figure 10
Figure 10. Figure 10: Perception-aware closed-loop control. V. CONCLUSION In this tutorial, we mainly focused on an anthology of generic optimization and computational tools for evalu￾ating the first characterization of the RDPF introduced in [2] for general i. i. d. source models and perc…

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

50 extracted references · 1 linked inside Pith

  1. [34]

    Universal rate- distortion-perception representations for lossy compression,

    G. Zhang, J. Qian, J. Chen, and A. Khisti, “Universal rate- distortion-perception representations for lossy compression,” Advances in Neural Information Processing Systems, vol. 34, pp. 11 517–11 529, 2021

    2021

  2. [1]

    Coding theorems for a discrete source with a fidelity criterion,

    C. E. Shannon, “Coding theorems for a discrete source with a fidelity criterion,”Institute of Radio Engineers, International Convention Record, vol. 7, pp. 325–350, 1959

    1959

  3. [2]

    Rethinking lossy compression: The rate-distortion-perception tradeoff,

    Y . Blau and T. Michaeli, “Rethinking lossy compression: The rate-distortion-perception tradeoff,” inInternational Conference on Machine Learning. PMLR, 2019, pp. 675–685

    2019

  4. [3]

    Semantics-empowered com- munication for networked intelligent systems,

    M. Kountouris and N. Pappas, “Semantics-empowered com- munication for networked intelligent systems,”IEEE Commun. Mag., vol. 59, no. 6, pp. 96–102, 2021

    2021

  5. [4]

    Goal-oriented and semantic communication in 6G AI-native networks: The 6G-GOALS approach,

    E. C. Strinati, P. Di Lorenzo, V . Sciancalepore, A. Aijaz, M. Kountouris, D. G¨und¨uz, P. Popovski, M. Sana, P. A. Stavrou, B. Soret, N. Cordeschi, S. Scardapane, M. Merluzzi, L. Zanzi, M. B. Renato, T. Quek, N. D. Pietro, O. Forceville, F. Costanzo, and P. Li, “Goal-oriented and semantic communication in 6G AI-native networks: The 6G-GOALS approach,” in...

    2024

  6. [5]

    A survey of recent results in networked control systems,

    J. P. Hespanha, P. Naghshtabrizi, and Y . Xu, “A survey of recent results in networked control systems,”Proc. IEEE, vol. 95, no. 1, pp. 138–162, 2007

    2007

  7. [6]

    Optimal neural compressors for the rate-distortion-perception tradeoff,

    E. Lei, H. Hassani, and S. S. Bidokhti, “Optimal neural compressors for the rate-distortion-perception tradeoff,” inThe Thirty-ninth Annual Conference on Neural Information Process- ing Systems, 2025

    2025

  8. [7]

    Rate–distortion–perception trade-off in information theory, generative models, and intelligent communications,

    X. Niu, B. Bai, N. Guo, W. Zhang, and W. Han, “Rate–distortion–perception trade-off in information theory, generative models, and intelligent communications,”Entropy, vol. 27, no. 4, 2025

    2025

Show all 50 references
  1. [8]

    Information compression in the AI era: Recent advances and future challenges,

    J. Chen, Y . Fang, A. Khisti, A. ¨Ozg¨ur, and N. Shlezinger, “Information compression in the AI era: Recent advances and future challenges,”IEEE J. Sel. Areas in Commun., vol. 43, no. 7, pp. 2333–2348, 2025

    2025

  2. [9]

    Introducing the perception-distortion tradeoff into the rate-distortion theory of general information sources,

    R. Matsumoto, “Introducing the perception-distortion tradeoff into the rate-distortion theory of general information sources,” IEICE Comm. Express, vol. 7, no. 11, pp. 427–431, 2018

    2018

  3. [10]

    Rate-distortion-perception tradeoff of variable-length source coding for general information sources,

    ——, “Rate-distortion-perception tradeoff of variable-length source coding for general information sources,”IEICE Comm. Express, vol. 8, no. 2, pp. 38–42, 2019

    2019

  4. [11]

    T. S. Han,Information-Spectrum Methods in Information The- ory. Springer-Verlag Berlin Heidelberg, 2003

    2003

  5. [12]

    A coding theorem for the rate- distortion-perception function,

    L. Theis and A. B. Wagner, “A coding theorem for the rate- distortion-perception function,” inInternational Conference of Learning Representations (ICLR): Neural Compression Work- shop, 2021, pp. 1–5

    2021

  6. [13]

    Strong functional representation lemma and applications to coding theorems,

    C. T. Li and A. E. Gamal, “Strong functional representation lemma and applications to coding theorems,”IEEE Trans. Inf. Theory, vol. 64, no. 11, pp. 6967–6978, 2018

    2018

  7. [14]

    On the rate-distortion-perception function,

    J. Chen, L. Yu, J. Wang, W. Shi, Y . Ge, and W. Tong, “On the rate-distortion-perception function,”IEEE Journal on Selected Areas in Information Theory, vol. 3, no. 4, pp. 664–673, 2022

    2022

  8. [15]

    Output constrained lossy source coding with limited common randomness,

    N. Saldi, T. Linder, and S. Y ¨uksel, “Output constrained lossy source coding with limited common randomness,”IEEE Trans. Inf. Theory, vol. 61, no. 9, pp. 4984–4998, 2015

    2015

  9. [16]

    The rate-distortion-perception tradeoff: The role of common randomness,

    A. B. Wagner, “The rate-distortion-perception tradeoff: The role of common randomness,” 2022. [Online]. Available: https://arxiv.org/abs/2202.04147

    2022 arXiv

  10. [17]

    Output-constrained lossy source coding with application to rate-distortion-perception theory,

    L. Xie, L. Li, J. Chen, and Z. Zhang, “Output-constrained lossy source coding with application to rate-distortion-perception theory,”IEEE Trans. Commun., vol. 73, no. 3, pp. 1801–1815, 2025

    2025

  11. [18]

    Gaus- sian rate–distortion–perception coding and entropy-constrained scalar quantization,

    L. Xie, L. Li, J. Chen, L. Yu, and Z. Zhang, “Gaus- sian rate–distortion–perception coding and entropy-constrained scalar quantization,”IEEE Trans. Commun., vol. 74, pp. 3298– 3312, 2026

    2026

  12. [19]

    Rate-distortion-perception theory for the quadratic Wasserstein space,

    X. Qu, J. Chen, L. Yu, and X. Xu, “Rate-distortion-perception theory for the quadratic Wasserstein space,”IEEE Trans. Inf. Theory, vol. 71, no. 11, pp. 8247–8261, 2025

    2025

  13. [20]

    On measures of entropy and information,

    A. R ´enyi, “On measures of entropy and information,”Proc. of the fourth Berkeley symposium on mathematical statistics and probability, volume 1: contributions to the theory of statistics, vol. 4, pp. 547–561, 1961

    1961

  14. [21]

    Information theory and statistics: A tutorial,

    I. Csisz ´ar and P. C. Shields, “Information theory and statistics: A tutorial,”Foundations and Trends® in Communications and Information Theory, vol. 1, no. 4, pp. 417–528, 2004

    2004

  15. [22]

    Onf-divergences: Integral representations, local behavior, and inequalities,

    I. Sason, “Onf-divergences: Integral representations, local behavior, and inequalities,”Entropy, vol. 20, no. 5, 2018

    2018

  16. [23]

    Information geometry and al- ternating minimization procedures,

    I. Csiszar and G. T ´usnady, “Information geometry and al- ternating minimization procedures,”Statistics and Decisions, Dedewicz, vol. 1, pp. 205–237, 1984

    1984

  17. [24]

    Yeung,Information theory and network coding

    R. Yeung,Information theory and network coding. Springer, 2008

    2008

  18. [25]

    On the convergence of the block nonlinear Gauss–Seidel method under convex constraints,

    L. Grippo and M. Sciandrone, “On the convergence of the block nonlinear Gauss–Seidel method under convex constraints,”Op- erations Research Letters, vol. 26, no. 3, pp. 127–136, 2000

    2000

  19. [26]

    An algorithm for computing the capacity of arbitrary discrete memoryless channels,

    S. Arimoto, “An algorithm for computing the capacity of arbitrary discrete memoryless channels,”IEEE Transactions on Information Theory, vol. 18, no. 1, pp. 14–20, 1972

    1972

  20. [27]

    Computation of channel capacity and rate- distortion functions,

    R. E. Blahut, “Computation of channel capacity and rate- distortion functions,”IEEE Trans. Inf. Theory, vol. 18, no. 4, pp. 460–473, 1972

    1972

  21. [28]

    Alternating minimization schemes for computing rate-distortion-perception functions with f-divergence perception constraints,

    G. Serra, P. A. Stavrou, and M. Kountouris, “Alternating minimization schemes for computing rate-distortion-perception functions with f-divergence perception constraints,”IEEE Trans. Inf. Theory, vol. 71, no. 11, pp. 9100–9115, 2025

    2025

  22. [29]

    R. L. Burden, J. D. Faires, and A. M. Burden,Numerical analysis. Cengage learning, 2015

    2015

  23. [30]

    A measure of asymptotic efficiency for tests of a hypothesis based on the sum of observations,

    H. Chernoff, “A measure of asymptotic efficiency for tests of a hypothesis based on the sum of observations,”Annals of Mathematical Statistics, vol. 23, no. 4, pp. 493–507, 1952

    1952

  24. [31]

    Amari,Differential geometric methods in statistics

    S. Amari,Differential geometric methods in statistics. New York: Springer Verlag, 1985

    1985

  25. [32]

    Analyzingα-divergence in Gaussian rate-distortion-perception theory,

    M. V . Sourla, G. Serra, P. A. Stavrou, and M. Kountouris, “Analyzingα-divergence in Gaussian rate-distortion-perception theory,” inIEEE 25 th International Workshop on Signal Pro- cessing Advances in Wireless Communications (SPAWC), 2024, pp. 856–860

    2024

  26. [33]

    On the computa- tion of the Gaussian rate-distortion-perception function,

    G. Serra, P. A. Stavrou, and M. Kountouris, “On the computa- tion of the Gaussian rate-distortion-perception function,”IEEE Journal on Selected Areas in Information Theory, vol. 5, pp. 314–330, 2024

    2024

  27. [35]

    Zamir,Lattice Coding for Signals and Networks: A Struc- tured Coding Approach to Quantization, Modulation and Mul- tiuser Information Theory

    R. Zamir,Lattice Coding for Signals and Networks: A Struc- tured Coding Approach to Quantization, Modulation and Mul- tiuser Information Theory. Cambridge University Press, 2014

    2014

  28. [36]

    Berger,Rate Distortion Theory: A Mathematical Basis for Data Compression

    T. Berger,Rate Distortion Theory: A Mathematical Basis for Data Compression. Prentice-Hall, 1971

    1971

  29. [37]

    Asymptotic reverse wa- terfilling algorithm of NRDF for certain classes of vec- tor Gauss–Markov processes,

    P. A. Stavrou and M. Skoglund, “Asymptotic reverse wa- terfilling algorithm of NRDF for certain classes of vec- tor Gauss–Markov processes,”IEEE Trans. Autom. Control, vol. 67, no. 6, pp. 3196–3203, 2022

    2022

  30. [38]

    Rate-distortion-perception tradeoff for Gaussian vector sources,

    J. Qian, S. Salehkalaibar, J. Chen, A. Khisti, W. Yu, W. Shi, Y . Ge, and W. Tong, “Rate-distortion-perception tradeoff for Gaussian vector sources,”IEEE J. Sel. Areas Inf. Theory, vol. 6, pp. 1–17, 2025

    2025

  31. [39]

    Copula-based estimation of continuous sources for a class of constrained rate-distortion functions,

    G. Serra, P. A. Stavrou, and M. Kountouris, “Copula-based estimation of continuous sources for a class of constrained rate-distortion functions,” inProc. IEEE Int. Symp. Inf. Theory, 2024, pp. 1089–1094

    2024

  32. [40]

    Generalizations of talagrand inequality for Sinkhorn distance using entropy power inequality,

    S. Wang, P. A. Stavrou, and M. Skoglund, “Generalizations of talagrand inequality for Sinkhorn distance using entropy power inequality,”Entropy, vol. 24, no. 2, 2022

    2022

  33. [41]

    Copula theory: An introduction,

    F. Durante and C. Sempi, “Copula theory: An introduction,” inCopula Theory and Its Applications. Berlin, Heidelberg: Springer Berlin Heidelberg, 2010, pp. 3–31

    2010

  34. [42]

    I-Divergence Geometry of Probability Distributions and Minimization Problems,

    I. Csisz ´ar, “I-Divergence Geometry of Probability Distributions and Minimization Problems,”The Annals of Probability, vol. 3, no. 1, pp. 146 – 158, 1975

    1975

  35. [43]

    Computation of rate-distortion-perception functions with Wasserstein barycenter,

    C. Chen, X. Niu, W. Ye, S. Wu, B. Bai, W. Chen, and S.-J. Lin, “Computation of rate-distortion-perception functions with Wasserstein barycenter,” inProc. IEEE Int. Symp. Inf. Theory, 2023, pp. 1074–1079

    2023

  36. [44]

    A convergent primal- dual algorithm for computing rate-distortion-perception func- tions,

    C. Chen, L. Chen, X. Niu, and H. Wu, “A convergent primal- dual algorithm for computing rate-distortion-perception func- tions,” 2025. 20

    2025

  37. [45]

    Computation and critical transitions of rate-distortion-perception functions with Wasserstein barycenter,

    C. Chen, X. Niu, W. Ye, H. Wu, and B. Bai, “Computation and critical transitions of rate-distortion-perception functions with Wasserstein barycenter,” 2024

    2024

  38. [46]

    El Gamal and Y .-H

    A. El Gamal and Y .-H. Kim,Network Information Theory. Cambridge University Press, 2011

    2011

  39. [47]

    A universal algorithm for sequential data compression,

    J. Ziv and A. Lempel, “A universal algorithm for sequential data compression,”IEEE Trans. Inf. Theory, vol. 23, no. 3, pp. 337–343, 1977

    1977

  40. [48]

    Simple Universal Lossy Data Com- pression Schemes derived from the Lempel-Ziv Algorithm,

    E.-H. Yang and J. Kieffer, “Simple Universal Lossy Data Com- pression Schemes derived from the Lempel-Ziv Algorithm,” IEEE Trans. Inf. Theory, vol. 42, no. 1, pp. 239–245, 1996

    1996

  41. [49]

    The Rate Distortion Function for a Class of Sources,

    D. J. Sakrison, “The Rate Distortion Function for a Class of Sources,”Information and Control, vol. 15, no. 2, pp. 165–195, 1969

    1969

  42. [50]

    On distributionally robust lossy source coding,

    G. Serra, P. A. Stavrou, and M. Kountouris, “On distributionally robust lossy source coding,” inProc. IEEE Inf. Theory Work- shop, 2025, pp. 1–6

    2025

Pith tools

Reviewed August 1, 2026 · model on record in the stance chip above.