pith. sign in

arxiv: 2605.15507 · v1 · pith:SQE5VREHnew · submitted 2026-05-15 · 💻 cs.IT · cs.AI· cs.LG· math.IT

PrismQuant: Rate-Distortion-Optimal Vector Quantization for Gaussian-Mixture Sources

Bumsu Park , Chanho Park , Youngmok Park , Namyoon Lee This is my paper

Pith reviewed 2026-05-19 15:16 UTC · model grok-4.3

classification 💻 cs.IT cs.AIcs.LGmath.IT
keywords rate-distortion optimizationvector quantizationGaussian mixture modeltransform codingreverse waterfillingKarhunen-Loeve transformscalar quantizationchannel state information
0
0 comments X

The pith

A single global reverse-waterfilling level governs the rate-distortion function for Gaussian-mixture sources.

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

The paper proves that once the active mixture component is known, the optimal bit allocation for a Gaussian-mixture source uses one common reverse-waterfilling level for every eigen-direction in every component. This finding implies that the total rate needed is only the entropy rate of the component label plus the conditional rate-distortion cost computed under that shared level. A reader would care because the result supplies both a theoretical characterization and a concrete coding method, PrismQuant, that comes within a vanishing gap of the fundamental limit while remaining simple to implement.

Core claim

The genie-aided conditional rate-distortion function of a Gaussian-mixture source is governed by a single global reverse-waterfilling level shared across all components and all eigenmodes. The paper constructs PrismQuant that transmits the component label losslessly at cost H(C)/n per dimension and then applies the component-matched Karhunen-Loeve transform followed by scalar quantization at levels set by the common water-filling threshold, achieving the conditional rate plus the label cost with a vanishing asymptotic gap.

What carries the argument

The single global reverse-waterfilling level shared across every mixture component and every eigenmode within those components.

Load-bearing premise

The source must be exactly a finite Gaussian mixture and the active component must be known perfectly so its label can be transmitted at exactly its entropy cost.

What would settle it

For a known two-component Gaussian mixture, compute the minimal rate required to achieve a target distortion and check whether it equals the shared reverse-waterfilling conditional rate plus the per-dimension component entropy.

Figures

Figures reproduced from arXiv: 2605.15507 by Bumsu Park, Chanho Park, Namyoon Lee, Youngmok Park.

Figure 1
Figure 1. Figure 1: Encoder and decoder of PRISMQUANT. The encoder selects the most likely mixture component, transmits its label losslessly, applies the component-matched KLT, and entropy-codes the active transform coefficients. The decoder inverts the pipeline through the shared dictionary θ = {(πc, µc, Rc)} K c=1. 4 The PrismQuant Codec PRISMQUANT is a practical hybrid lossless–lossy coding framework for the Gaussian-mixtu… view at source ↗
Figure 2
Figure 2. Figure 2: Left: Huffman code for C with prior ( 1 2 , 1 4 , 1 8 , 1 8 ), H(π) = 1.75 bits/symbol. Right: TC (top), PRISMQUANT (middle), and k-means VQ (bottom) on a four-component mixture. PRISMQUANT (middle) sends the label and then runs a component-matched KLT plus scalar quanti￾zation, so the per-component grids align with the local covariance ellipses; an unstructured k-means VQ [20] (bottom) recovers similar Vo… view at source ↗
Figure 3
Figure 3. Figure 3: Reverse waterfilling for a Gaussian-mixture source ( [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Effect of mixture order K at fixed n = 32. Operational (Genie) closely tracks the theoretical upper bound, validating Theorem 1; Operational (MAP) nearly coincides with Genie as K → ∞. MAP label error in high dimension. Practical PRISMQUANT replaces the oracle label with a hard MAP decision Cˆ(x) = arg maxc πcN (x; µc, Rc). The standard Bhattacharyya union bound gives P MAP e ≤ P c<j √πcπj ρ (n) c,j with ρ… view at source ↗
Figure 5
Figure 5. Figure 5: Spatial visualization of PRISMQUANT’s clustering on the DeepMIMO O1 dataset. (a) The outdoor deployment map used for CSI generation. (b)–(d) MAP component assignments—obtained from the EM-fitted Gaussian-mixture dictionary—plotted at each user’s location, with varying K. Pixels sharing the same color correspond to CSI samples assigned to the same component. 0.0 0.1 0.2 0.3 0.4 0.5 Rate (bits/complex dimens… view at source ↗
Figure 6
Figure 6. Figure 6: RD performance on the DeepMIMO O1 dataset. (a) [PITH_FULL_IMAGE:figures/full_fig_p009_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Synthetic Gaussian-mixture RD comparison. [PITH_FULL_IMAGE:figures/full_fig_p017_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Effect of source dimension n at fixed mixture order K = 16. As n increases, the gap between the theoretical lower and upper bounds progressively shrinks, consistent with Corollary 1. Operational (Genie) closely tracks the theoretical upper bound, while Operational (MAP) even outperforms Operational (Genie) scheme when n = 4. 0 2 4 6 8 10 12 14 16 Rate (bits/complex dimension) 80 70 60 50 40 30 20 10 0 NMSE… view at source ↗
Figure 9
Figure 9. Figure 9: Effect of block length n on DeepMIMO at varying mixture order K. Together with [PITH_FULL_IMAGE:figures/full_fig_p018_9.png] view at source ↗
read the original abstract

For a Gaussian source under mean-squared error (MSE), classical transform coding is rate--distortion (RD) optimal: the Karhunen--Loeve transform (KLT) diagonalizes the covariance, reverse waterfilling allocates the bits, and scalar quantization closes the loop. This elegant story breaks down for multimodal sources, where no single covariance can capture heterogeneous local geometries, and the RD function loses its closed form. We revisit this problem through Gaussian-mixture sources and develop a constructive RD theory for them. Our key finding is that the mixture structure incurs only a component label cost. Conditioned on the active mixture component, each branch is Gaussian; the challenge is allocating bits across heterogeneous branches. We prove that the genie-aided conditional RD function is governed by a single global reverse-waterfilling level shared across all components and eigenmodes. Building on this result, we introduce PrismQuant, which transmits the component label losslessly and encodes the residual using the component-matched KLT, followed by scalar quantization, achieving a rate of H(C)/n bits per source dimension of the converse, with a vanishing asymptotic gap. We further develop a practical implementation based on EM-driven Gaussian-mixture learning, component-adaptive KLTs, and entropy-constrained scalar quantization (ECSQ). Experiments on synthetic Gaussian mixtures show that PrismQuant closely approaches the theoretical RD bound, while experiments on real-world channel-state-information (CSI) data demonstrate competitive or superior performance compared with transformer-based learned codecs at more than one order of magnitude smaller model size.

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 paper develops a constructive rate-distortion theory for finite Gaussian-mixture sources under MSE. It proves that the genie-aided conditional RD function is governed by a single global reverse-waterfilling level shared across all mixture components and their eigenmodes. PrismQuant is introduced to transmit the component label losslessly at rate H(C)/n per dimension and then apply the component-matched KLT followed by scalar quantization, achieving a rate within H(C)/n of the converse with a vanishing asymptotic gap. A practical implementation learns the GMM via EM, uses component-adaptive KLTs, and employs entropy-constrained scalar quantization; experiments on synthetic GMMs approach the theoretical bound while CSI data experiments show competitive performance against transformer codecs at far smaller model size.

Significance. If the central result holds, the work cleanly extends classical transform coding to multimodal sources by isolating the mixture overhead to the label entropy term and supplying an explicit, low-complexity construction that approaches the information-theoretic limit. The practical codec's small footprint and strong CSI results indicate immediate relevance to quantization and compression in wireless and sensing applications.

major comments (3)
  1. [§3] §3 (Global reverse-waterfilling derivation): the proof equalizes the Lagrange multiplier across eigenmodes of every component under the assumption that a genie reveals the active component C before quantization; the manuscript must clarify whether and how this equalization survives when the encoder must infer C from X, since any positive misclassification probability applies the wrong covariance and quantizer.
  2. [§4] §4 (PrismQuant construction and gap analysis): the claim of a vanishing asymptotic gap to the converse rests on lossless transmission of the true component label; when mixture components overlap, the encoder's inference of C is imperfect and the resulting excess distortion from mismatched KLT/ECSQ may not vanish, requiring either a bound on the misclassification contribution or an explicit statement that the gap analysis assumes perfect identification.
  3. [Experimental section] Experimental section (synthetic GMM results): the reported approach to the theoretical bound is promising, yet the absence of error bars, number of Monte-Carlo trials, or variance across random seeds leaves open whether the observed gap is statistically consistent with vanishing or merely small for the tested dimensions.
minor comments (2)
  1. [Notation] Notation for the per-dimension label cost H(C)/n should be introduced once and used consistently; the abstract and main text occasionally switch between “bits per source dimension” and “per dimension” without explicit definition of n.
  2. [Figures] Figure captions for the CSI experiments should state the exact model sizes of the transformer baselines and the number of parameters in PrismQuant for direct comparison.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the detailed and constructive review of our manuscript. The comments highlight important points regarding the distinction between genie-aided and practical settings, as well as the need for more rigorous statistical reporting in experiments. We address each major comment below and outline the revisions we intend to make in the next version of the paper.

read point-by-point responses

  1. Referee: [§3] §3 (Global reverse-waterfilling derivation): the proof equalizes the Lagrange multiplier across eigenmodes of every component under the assumption that a genie reveals the active component C before quantization; the manuscript must clarify whether and how this equalization survives when the encoder must infer C from X, since any positive misclassification probability applies the wrong covariance and quantizer.

    Authors: The analysis in §3 derives the conditional rate-distortion function under the genie-aided assumption that the active component C is known. The equalization of the Lagrange multiplier across all eigenmodes and components follows from this conditional setting. We will revise the manuscript to explicitly clarify this assumption and explain that PrismQuant uses an estimate of C, with the theoretical bound holding in the genie-aided case. The practical performance is evaluated separately in the experiments. revision: yes

  2. Referee: [§4] §4 (PrismQuant construction and gap analysis): the claim of a vanishing asymptotic gap to the converse rests on lossless transmission of the true component label; when mixture components overlap, the encoder's inference of C is imperfect and the resulting excess distortion from mismatched KLT/ECSQ may not vanish, requiring either a bound on the misclassification contribution or an explicit statement that the gap analysis assumes perfect identification.

    Authors: We agree that the vanishing gap analysis in §4 is predicated on the lossless transmission of the true component label, which corresponds to the H(C)/n rate overhead in the theoretical construction. When components overlap significantly, perfect inference is not always possible, and mismatched quantization can introduce additional distortion. We will add an explicit statement in the revised manuscript clarifying that the gap analysis assumes perfect component identification (as in the genie-aided bound), and note that for the practical implementation, the performance approaches the bound when the mixture components are sufficiently separated or when the classification accuracy is high. A full bound on the misclassification term is left for future work as it would require additional assumptions on the component separation. revision: partial

  3. Referee: Experimental section (synthetic GMM results): the reported approach to the theoretical bound is promising, yet the absence of error bars, number of Monte-Carlo trials, or variance across random seeds leaves open whether the observed gap is statistically consistent with vanishing or merely small for the tested dimensions.

    Authors: We appreciate this observation. To address it, we will rerun the synthetic GMM experiments with multiple Monte-Carlo trials (specifically, 20 independent trials with different random seeds for initialization and data generation) and include error bars representing the standard deviation in the revised manuscript. This will confirm that the gaps are small and consistent with the expected vanishing behavior as dimension increases. revision: yes

Circularity Check

0 steps flagged

No significant circularity; derivation uses standard conditional Gaussian RD theory

full rationale

The paper's central result derives the single global reverse-waterfilling level for the genie-aided conditional RD function directly from classical Gaussian rate-distortion theory applied separately to each mixture component after conditioning on the active label. The PrismQuant construction then explicitly transmits the component label losslessly at cost H(C)/n and applies the matched KLT plus scalar quantization to the residual; this is a constructive scheme whose rate expression follows from the preceding independent derivation rather than redefining or fitting the target quantity to itself. No step reduces the claimed vanishing asymptotic gap to a fitted parameter, self-citation chain, or ansatz smuggled from prior work by the same authors. The practical EM-based implementation is presented separately from the theoretical converse and does not alter the first-principles status of the RD proof.

Axiom & Free-Parameter Ledger

1 free parameters · 2 axioms · 0 invented entities

The central claim rests on the assumption that the source is exactly a finite Gaussian mixture and on the standard single-Gaussian RD theory applied conditionally; no new physical constants or entities are introduced.

free parameters (1)
  • number of mixture components
    Chosen during EM learning to fit the data; affects the label entropy H(C) and the per-component transforms.
axioms (2)
  • domain assumption The source distribution is exactly a finite Gaussian mixture model
    Invoked throughout the RD analysis and the construction of PrismQuant.
  • standard math Standard reverse waterfilling and KLT are RD-optimal for each conditional Gaussian component
    Used to obtain the global waterfilling level once the component is known.

pith-pipeline@v0.9.0 · 5825 in / 1489 out tokens · 49008 ms · 2026-05-19T15:16:10.431011+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

40 extracted references · 40 canonical work pages · 2 internal anchors

  1. [1]

    Claude E. Shannon. Coding theorems for a discrete source with a fidelity criterion.IRE National Convention Record, 7(4):142–163, 1959

    work page 1959

  2. [2]

    Prentice-Hall, Englewood Cliffs, NJ, 1971

    Toby Berger.Rate distortion theory: A mathematical basis for data compression. Prentice-Hall, Englewood Cliffs, NJ, 1971

    work page 1971

  3. [3]

    Cover and Joy A

    Thomas M. Cover and Joy A. Thomas.Elements of information theory. Wiley–Interscience, New York, NY , 2nd edition, 2006

    work page 2006

  4. [4]

    Vivek K. Goyal. Theoretical foundations of transform coding.IEEE Signal Procesing Magazine, 18(5): 9–21, 2001

    work page 2001

  5. [5]

    Bishop.Pattern recognition and machine learning

    Christopher M. Bishop.Pattern recognition and machine learning. Springer, New York, NY , 2006

    work page 2006

  6. [6]

    T. L. Marzetta. Noncooperative cellular wireless with unlimited numbers of base station antennas.IEEE Transactions on Wireless Communications, 9(11):3590–3600, 2010

    work page 2010

  7. [7]

    D. J. Love, R. W. Heath, Jr., V . K. N. Lau, D. Gesbert, B. D. Rao, and M. Andrews. An overview of limited feedback in wireless communication systems.IEEE Journal on Selected Areas in Communications, 26(8): 1341–1365, 2008

    work page 2008

  8. [8]

    R. W. Heath, Jr. and A. Lozano.Foundations of MIMO Communication. Cambridge Univ. Press, 2016

    work page 2016

  9. [9]

    Molisch, and Pierre De Doncker

    Lingfeng Liu, Claude Oestges, Juho Poutanen, Francois Quitin, Katsuyuki Haneda, Pekka Vainikainen, Fredrik Tufvesson, Andreas F. Molisch, and Pierre De Doncker. The COST 2100 MIMO channel model. IEEE Wireless Communications, 19(6):92–99, 2012

    work page 2012

  10. [10]

    Wen, W.-T

    C.-K. Wen, W.-T. Shih, and S. Jin. Deep learning for massive MIMO CSI feedback.IEEE Wireless Commun. Lett., 7(5):748–751, 2018

    work page 2018

  11. [11]

    B. Park, H. Do, and N. Lee. Transformer-based nonlinear transform coding for multi-rate CSI compression in MIMO-OFDM systems. InProceedings of IEEE International Conference on Communications (ICC), 2025

    work page 2025

  12. [12]

    On the entropy of general mixture distributions, 2026

    Namyoon Lee. On the entropy of general mixture distributions, 2026. arXiv:2602.15303

    work page arXiv 2026

  13. [13]

    Effros and P

    M. Effros and P. A. Chou. Weighted universal transform coding: Universal image compression with the Karhunen–Loève transform. InProceedings of IEEE International Conference on Image Processing (ICIP), 1995

    work page 1995

  14. [14]

    DeepMIMO: A Generic Deep Learning Dataset for Millimeter Wave and Massive MIMO Applications

    Ahmed Alkhateeb. DeepMIMO: A generic deep learning dataset for millimeter wave and massive MIMO applications, 2019. arXiv:1902.06435

    work page internal anchor Pith review Pith/arXiv arXiv 2019

  15. [15]

    Robert M. Gray. Conditional rate–distortion theory. Technical report, Stanford Electronics Laboratories, Stanford University, 1972

    work page 1972

  16. [16]

    A. D. Wyner and J. Ziv. The rate–distortion function for source coding with side information at the decoder. IEEE Transactions on Information Theory, 22(1):1–10, 1976

    work page 1976

  17. [17]

    Heegard and T

    C. Heegard and T. Berger. Rate distortion when side information may be absent.IEEE Transactions on Information Theory, 31(6):727–734, 1985

    work page 1985

  18. [18]

    E. A. Riskin and R. M. Gray. A greedy tree growing algorithm for the design of variable rate vector quantizers.IEEE Transactions on Signal Processing, 39(11):2500–2507, 1991

    work page 1991

  19. [19]

    Effros, P

    M. Effros, P. A. Chou, and R. M. Gray. Variable dimension weighted universal vector quantization and noiseless coding. InProceedings of IEEE Data Compression Conference (DCC), 1994

    work page 1994

  20. [20]

    Linde, A

    Y . Linde, A. Buzo, and R. M. Gray. An algorithm for vector quantizer design.IEEE Transactions on Communications, 28(1):84–95, 1980

    work page 1980

  21. [21]

    Hedelin and J

    P. Hedelin and J. Skoglund. Vector quantization based on Gaussian mixture models.IEEE Transactions on Speech and Audio Processing, 8(4):385–401, 2000

    work page 2000

  22. [22]

    J. Han, A. Saxena, and K. Rose. Towards jointly optimal spatial prediction and adaptive transform in video/image coding. InProc. IEEE Int. Conf. Acoust., Speech, Signal Process. (ICASSP), pages 726–729, Mar. 2010. 11

    work page 2010

  23. [23]

    J. Han, A. Saxena, V . Melkote, and K. Rose. Jointly optimized spatial prediction and block transform for video and image coding.IEEE Trans. Image Process., 21(4):1874–1884, Apr. 2012

    work page 2012

  24. [24]

    Saxena and F

    A. Saxena and F. C. Fernandes. DCT/DST-based transform coding for intra prediction in image/video coding.IEEE Trans. Image Process., 22(10):3974–3981, Oct. 2013

    work page 2013

  25. [25]

    X. Zhao, J. Chen, M. Karczewicz, A. Said, and V . Seregin. Enhanced multiple transform for video coding. InProc. Data Compression Conf. (DCC), pages 73–82, Mar. 2016

    work page 2016

  26. [26]

    Bross, Y .-K

    B. Bross, Y .-K. Wang, Y . Ye, S. Liu, J. Chen, G. J. Sullivan, and J.-R. Ohm. Overview of the Versatile Video Coding (VVC) standard and its applications.IEEE Trans. Circuits Syst. Video Technol., 31(10): 3736–3764, Oct. 2021

    work page 2021

  27. [27]

    Variational image compression with a scale hyperprior

    Johannes Ballé, David Minnen, Saurabh Singh, Sung Jin Hwang, and Nick Johnston. Variational image compression with a scale hyperprior. InProceedings of International Conference on Learning Representa- tions (ICLR), 2018

    work page 2018

  28. [28]

    Learned image compression with dis- cretized gaussian mixture likelihoods and attention modules

    Zhengxue Cheng, Heming Sun, Masaru Takeuchi, and Jiro Katto. Learned image compression with dis- cretized gaussian mixture likelihoods and attention modules. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2020

    work page 2020

  29. [29]

    Neural estimation of the rate-distortion function with applications to operational source coding.IEEE Journal on Selected Areas in Information Theory, 3 (4):674–686, 2022

    Eric Lei, Hamed Hassani, and Shirin Saeedi Bidokhti. Neural estimation of the rate-distortion function with applications to operational source coding.IEEE Journal on Selected Areas in Information Theory, 3 (4):674–686, 2022

    work page 2022

  30. [30]

    TurboQuant: Online Vector Quantization with Near-optimal Distortion Rate

    Amir Zandieh, Majid Daliri, Majid Hadian, and Vahab Mirrokni. TurboQuant: Online vector quantization with near-optimal distortion rate, 2025. arXiv:2504.19874

    work page internal anchor Pith review arXiv 2025

  31. [31]

    A. P. Dempster, N. M. Laird, and D. B. Rubin. Maximum likelihood from incomplete data via the EM algorithm.Journal on Royal Statistical Society: Series B (Methodological), 39(1):1–22, 1977

    work page 1977

  32. [32]

    Channel estimation for massive MIMO using Gaussian-mixture bayesian learning.IEEE Transactions on Wireless Communications, 14(3): 1356–1368, 2015

    Chao-Kai Wen, Shi Jin, Kai-Kit Wong, Jung-Chieh Chen, and Pangan Ting. Channel estimation for massive MIMO using Gaussian-mixture bayesian learning.IEEE Transactions on Wireless Communications, 14(3): 1356–1368, 2015

    work page 2015

  33. [33]

    Downlink CSIT under compressed feedback: Joint versus separate source-channel coding.IEEE Transactions on Wireless Communications, 24(10):8429–8444, 2025

    Yi Song, Tianyu Yang, Mahdi Barzegar Khalilsarai, and Giuseppe Caire. Downlink CSIT under compressed feedback: Joint versus separate source-channel coding.IEEE Transactions on Wireless Communications, 24(10):8429–8444, 2025

    work page 2025

  34. [34]

    Nguyen, Marcel Worring, and Arnold W

    Hieu T. Nguyen, Marcel Worring, and Arnold W. M. Smeulders. A study of Gaussian mixture models of color and texture features for image classification and segmentation.Pattern Recognition, 38(11): 2005–2018, 2005

    work page 2005

  35. [35]

    From learning models of natural image patches to whole image restoration

    Daniel Zoran and Yair Weiss. From learning models of natural image patches to whole image restoration. InProceedings of IEEE International Conference on Computer Vision (ICCV), 2011

    work page 2011

  36. [36]

    Multi-facet clustering variational autoencoders.Advances in Neural Information Processing Systems, 34: 8676–8690, 2021

    Fabian Falck, Haoting Zhang, Matthew Willetts, George Nicholson, Christopher Yau, and Chris C Holmes. Multi-facet clustering variational autoencoders.Advances in Neural Information Processing Systems, 34: 8676–8690, 2021

    work page 2021

  37. [37]

    Subspace gaussian mixture models for speech recognition

    Daniel Povey, Lukas Burget, Mohit Agarwal, Pinar Akyazi, Fan Kai, Arnab Ghoshal, Ondrej Glembek, Nagendra Goel, Martin Karafiat, Ariya Rastrow, et al. Subspace gaussian mixture models for speech recognition. InProceedings of IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2010

    work page 2010

  38. [38]

    Gray.Vector quantization and signal compression

    Allen Gersho and Robert M. Gray.Vector quantization and signal compression. Springer, Boston, MA, 1992

    work page 1992

  39. [39]

    A Markovian model-driven deep learning framework for massive MIMO CSI feedback.IEEE Transactions on Wireless Communications, 21(2):1214–1228, 2022

    Zhenyu Liu, Mason del Rosario, and Zhi Ding. A Markovian model-driven deep learning framework for massive MIMO CSI feedback.IEEE Transactions on Wireless Communications, 21(2):1214–1228, 2022. 12 A Proof of Theorem 1 (single global water level) The conditional benchmark (4) is a separable convex program over {Dc} subject to a single linear constraint. For...

    work page 2022

  40. [40]

    documents its current form in VVC. The library is a small set offixed trigonometric transforms (DCT-II together with DST-VII and DCT-VIII variants), the branch is chosen by an encoder-side rate–distortion search and signaled with a few context-coded bits, and the bit allocation is governed by a per-block quantization parameter from the standard. AMT thus ...

    work page 2048