pith. sign in

arxiv: 2605.18659 · v1 · pith:XSX3S6K3new · submitted 2026-05-18 · 🌌 astro-ph.CO

Faster CMB lensing with control variates

Toshiya Namikawa , Blake D. Sherwin This is my paper

Pith reviewed 2026-05-20 09:02 UTC · model grok-4.3

classification 🌌 astro-ph.CO
keywords CMB lensinglensing power spectrumrealization-dependent biascontrol variatessimulation-based methodscosmic microwave backgroundcosmological parameter estimation
0
0 comments X

The pith

Differencing realistic and isotropic CMB simulations cuts the computational cost of lensing power spectrum bias calculations by a factor of five.

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

The paper presents a method to accelerate the calculation of realization-dependent bias, which is a major computational bottleneck when measuring the cosmic microwave background lensing power spectrum. The approach works by subtracting a correlated estimate from isotropic simulations, where the bias has a known analytical form, from estimates based on fully realistic masked simulations. This differencing reduces variance while preserving unbiased results and can be added to existing analysis codes. A sympathetic reader would care because the technique lowers the total cost of a full lensing power spectrum measurement by roughly a factor of five at typical next-generation noise levels, freeing resources for larger data sets or more precise analyses.

Core claim

We present a new method for fast computation of the realization-dependent bias, a major computational bottleneck in measurements of the cosmic microwave background lensing power spectrum. The method accelerates the bias calculation by differencing two correlated estimates: one based on fully realistic masked simulations and the other on isotropic simulations, for which the bias is analytically tractable. We show that our algorithm reduces the total computational cost of a lensing power spectrum measurement by approximately a factor of five for Atacama Cosmology Telescope- or Simons Observatory-like noise levels, or by a factor of three if current anisotropic filtering methods are leftunch.

What carries the argument

Control variate formed by subtracting an isotropic-simulation bias estimate (analytically tractable) from a realistic masked-simulation estimate to reduce variance of the realization-dependent bias term.

If this is right

  • A full lensing power spectrum measurement requires roughly one-fifth as many simulations at typical upcoming experiment noise levels.
  • The same method yields a factor-of-three cost reduction even when anisotropic filtering is kept unchanged.
  • The algorithm is simple enough to drop into existing CMB lensing pipelines without major code changes.
  • Fewer simulations needed means either lower total compute or the ability to process larger sky fractions or more frequency channels with fixed resources.

Where Pith is reading between the lines

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

  • The same control-variate idea could be tested on other quadratic estimators or on lensing reconstruction from future experiments with different mask geometries.
  • If the correlation between realistic and isotropic estimates remains high at lower noise levels, the method may deliver even larger speed-ups for CMB-S4 or similar surveys.
  • Combining this differencing with other variance-reduction techniques already in use could compound the savings beyond the reported factors.

Load-bearing premise

The estimates from fully realistic masked simulations and isotropic simulations are sufficiently correlated that their difference yields a large variance reduction, while the bias for the isotropic case remains analytically tractable.

What would settle it

Measure the variance of the bias estimator on a set of end-to-end masked simulations with ACT- or SO-like noise and confirm whether the reduction in required simulation count reaches approximately five relative to the standard method.

Figures

Figures reproduced from arXiv: 2605.18659 by Blake D. Sherwin, Toshiya Namikawa.

Figure 1
Figure 1. Figure 1: FIG. 1. The Planck PR3 and ACTDR6 masks used for the lensing analysis, shown in Galactic coordinates. For the Planck [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: FIG. 2. Comparison of the control-variate estimator for the data–simulation cross term in Eq. ( [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: FIG. 3 [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: FIG. 4. Same as the left panel of Fig [PITH_FULL_IMAGE:figures/full_fig_p007_4.png] view at source ↗
read the original abstract

We present a new method for fast computation of the realization-dependent bias, a major computational bottleneck in measurements of the cosmic microwave background (CMB) lensing power spectrum. The method accelerates the bias calculation by differencing two correlated estimates: one based on fully realistic masked simulations and the other on isotropic simulations, for which the bias is analytically tractable. We show that our algorithm reduces the total computational cost of a lensing power spectrum measurement by approximately a factor of five for Atacama Cosmology Telescope- or Simons Observatory-like noise levels, or by a factor of three if current anisotropic filtering methods are left unchanged. Owing to its simplicity, the method can be readily implemented in existing CMB lensing analysis pipelines.

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

2 major / 2 minor

Summary. The manuscript introduces a control-variate algorithm to accelerate computation of the realization-dependent bias in CMB lensing power-spectrum measurements. One estimate is obtained from fully realistic masked simulations; a second, correlated estimate is obtained from isotropic simulations whose bias admits an analytic expression. Subtracting the two yields a low-variance bias correction. The authors report that the method reduces the total computational cost of a lensing power-spectrum measurement by a factor of approximately five for ACT- or SO-like noise levels (or a factor of three if anisotropic filtering is retained).

Significance. If the reported speed-up is robust, the technique would materially lower the simulation budget required for high-precision lensing analyses in Stage-3 and Stage-4 CMB surveys. The method’s simplicity and compatibility with existing pipelines are practical advantages.

major comments (2)
  1. [Abstract and §4] Abstract and §4: the factor-of-five (or factor-of-three) cost reduction is the central performance claim. This reduction requires the correlation coefficient between the masked-realistic and isotropic estimates to satisfy r ≳ 0.95 at the multipoles that dominate the lensing spectrum. The manuscript does not report the measured correlation coefficient, the effective number of simulations saved, or the ratio Var(Δ)/Var(realistic) in any figure or table. Without these diagnostics the numerical claim cannot be independently verified.
  2. [§3.2] §3.2: the analytic bias expression for the isotropic case is invoked after masking and anisotropic filtering have been applied to the realistic simulations. The text does not quantify how strongly these operations decorrelate the two fields at the scales relevant to the lensing power spectrum. A short appendix or figure showing the scale-dependent correlation would directly address the skeptic’s concern.
minor comments (2)
  1. [Figure 2] Figure 2 caption: the error bars shown are the final lensing-power-spectrum uncertainties; adding a companion panel or table that isolates the variance reduction achieved by the control variate would make the performance gain transparent.
  2. [Methods] Notation: the symbol used for the isotropic bias correction is introduced without an explicit equation number in the methods section; adding an equation label would improve traceability.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their careful reading of the manuscript and for the constructive comments. We agree that the requested diagnostics will strengthen the presentation and allow independent verification of the performance claims. We have revised the manuscript to incorporate these elements and address each major comment below.

read point-by-point responses

  1. Referee: [Abstract and §4] Abstract and §4: the factor-of-five (or factor-of-three) cost reduction is the central performance claim. This reduction requires the correlation coefficient between the masked-realistic and isotropic estimates to satisfy r ≳ 0.95 at the multipoles that dominate the lensing spectrum. The manuscript does not report the measured correlation coefficient, the effective number of simulations saved, or the ratio Var(Δ)/Var(realistic) in any figure or table. Without these diagnostics the numerical claim cannot be independently verified.

    Authors: We agree that explicit reporting of these quantities would improve verifiability. In the revised manuscript we have added a new figure in §4 that shows the measured correlation coefficient r(ℓ) between the two estimates; r exceeds 0.95 across the multipole range that dominates the lensing power spectrum. We also include a table reporting the effective number of simulations saved and the ratio Var(Δ)/Var(realistic) for ACT- and SO-like noise levels, directly confirming the stated factor-of-five (or factor-of-three) cost reduction. revision: yes

  2. Referee: [§3.2] §3.2: the analytic bias expression for the isotropic case is invoked after masking and anisotropic filtering have been applied to the realistic simulations. The text does not quantify how strongly these operations decorrelate the two fields at the scales relevant to the lensing power spectrum. A short appendix or figure showing the scale-dependent correlation would directly address the skeptic’s concern.

    Authors: We appreciate the suggestion. The revised manuscript now includes a new figure in §3.2 (with supporting discussion) that displays the scale-dependent correlation coefficient between the masked, anisotropically filtered realistic estimates and the isotropic estimates. The figure shows that the correlation remains high (r ≳ 0.9) at the scales that contribute most to the lensing spectrum, justifying continued use of the analytic isotropic bias expression. The text in §3.2 has been updated to reference this result. revision: yes

Circularity Check

0 steps flagged

No significant circularity; method is self-contained

full rationale

The core algorithm subtracts an isotropic simulation estimate (with analytically tractable bias) from a realistic masked simulation estimate to reduce variance in the realization-dependent bias correction. This relies on the physical correlation between the two estimates and the standard property that isotropic cases admit analytic bias expressions—both external to the result itself and verifiable via simulations. No self-definitional steps, fitted inputs renamed as predictions, or load-bearing self-citations appear in the derivation chain. The factor-of-five cost reduction follows from the control variate construction and measured correlation, without reducing to the inputs by construction.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

Review performed on abstract only; full manuscript may contain additional parameters or assumptions not visible here.

axioms (2)
  • domain assumption The bias for isotropic simulations is analytically tractable.
    Explicitly invoked in the abstract as the reason the control variate works.
  • domain assumption Realistic masked and isotropic simulation estimates are highly correlated.
    Required for the variance reduction that produces the claimed computational saving.

pith-pipeline@v0.9.0 · 5640 in / 1356 out tokens · 41581 ms · 2026-05-20T09:02:27.753249+00:00 · methodology

discussion (0)

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

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

59 extracted references · 59 canonical work pages · 25 internal anchors

  1. [1]

    construction in which the conventional RDN0 estimator is combined with RDN0-like control variates computed from statistically isotropic lensed CMB simulations

    We also show cases with weaker filtering,|ℓ x|<45 and |ℓy|<25, and with Planck-like information used to fill the missing modes up toℓ≤2000. construction in which the conventional RDN0 estimator is combined with RDN0-like control variates computed from statistically isotropic lensed CMB simulations. We found that the isotropic control variate is highly cor...

    work page 2000

  2. [2]

    RDN0 in the flat-sky approximation We first summarize the expression for RDN0 in the flat-sky approximation. The data–simulation cross term in the flat-sky RDN0 is given by (2π)2Nf L ≡(A f L)2 Z d2ℓ (2π)2 Z d2ℓ′ (2π)2 fℓ,Lfℓ′,L⟨ ¯Θℓ ¯Θ∗ ℓ′⟩ ¯Θℓ ¯Θ∗ ℓ′ + (3 perms.),(A1) whereA f L is the normalization of the flat-sky estimator, ¯Θℓ denotes the Fourier mode...

    work page

  3. [3]

    The analysis mask is then applied, followed by a Fourier transform

    A flat-sky control variate for RDN0 For ACT-like simulations, we assume that the observed temperature map is first projected onto a CAR pixelization. The analysis mask is then applied, followed by a Fourier transform. We further apply the Fourier-space filter to 9 the temperature modes, transform the filtered map back to real space, and finally compute th...

    work page

  4. [4]

    Blanchard and J

    A. Blanchard and J. SchneiderAstron. Astrophys.184(1987) 1–6

    work page 1987

  5. [5]

    Weak Gravitational Lensing of the CMB

    A. Lewis and A. ChallinorPhys. Rep.429(2006) 1–65,astro-ph/0601594

    work page internal anchor Pith review Pith/arXiv arXiv 2006

  6. [6]

    S. Das, B. D. Sherwin, P. Aguirre, J. W. Appel, J. R. Bond, C. S. Carvalho, M. J. Devlin, J. Dunkley, R. D¨ unneret al., Phys. Rev. Lett.107(2011), no. 2 021301,arXiv:1103.2124

    work page internal anchor Pith review Pith/arXiv arXiv 2011

  7. [7]

    B. D. Sherwinet al., Phys. Rev. Lett.107(2011) 021302,arXiv:1105.0419

    work page internal anchor Pith review Pith/arXiv arXiv 2011

  8. [8]

    B. D. Sherwin, A. van Engelen, N. Sehgal, M. Madhavacheril, G. E. Addison, S. Aiola, R. Allison, N. Battaglia, D. T. Beckeret al., Phys. Rev. D95(2017) 123529,arXiv:1611.09753

    work page internal anchor Pith review Pith/arXiv arXiv 2017

  9. [9]

    F. J. Quet al., Astrophys. J.962(2024), no. 2 112,arXiv:2304.05202

    work page internal anchor Pith review arXiv 2024

  10. [10]

    MacCrannet al.(ACT), Astrophys

    N. MacCrannet al., Astrophys. J.966(2024), no. 1 138,arXiv:2304.05196

    work page arXiv 2024

  11. [11]

    Abril-Cabezaset al.,arXiv:2511.10620

    I. Abril-Cabezaset al.,arXiv:2511.10620. [9]Bicep2/Keck ArrayCollaborationAstrophys. J.833(2016) 228,arXiv:1606.01968. [10]Bicep2/Keck ArrayCollaborationAstrophys. J.949(2023), no. 2 43,arXiv:2210.08038. [11]PlanckCollaborationAstron. Astrophys.571(2014) A17,arXiv:1303.5077. [12]PlanckCollaborationAstron. Astrophys.594(2015) A15,arXiv:1502.01591

    work page arXiv 2016

  12. [12]

    CMB lensing from Planck PR4 maps

    J. Carron, M. Mirmelstein, and A. LewisJ. Cosmol. Astropart. Phys.09(2022) 039,arXiv:2206.07773. [14]POLARBEAR CollaborationPhys. Rev. Lett.113(2014) 021301,arXiv:1312.6646. [15]POLARBEAR CollaborationAstrophys. J.893(2020) 85,arXiv:1911.10980

    work page internal anchor Pith review Pith/arXiv arXiv 2022

  13. [13]

    A measurement of gravitational lensing of the microwave background using South Pole Telescope data

    A. van Engelen, R. Keisler, O. Zahn, K. A. Aird, B. A. Benson, L. E. Bleem, J. E. Carlstrom, C. L. Chang, H. M. Cho et al., Astrophys. J.756(2012), no. 2 142,arXiv:1202.0546

    work page internal anchor Pith review Pith/arXiv arXiv 2012

  14. [14]

    K. T. Story, D. Hanson, P. A. R. Ade, K. A. Aird, J. E. Austermann, J. A. Beall, A. N. Bender, B. A. Benson, L. E. Bleemet al., Astrophys. J.810(aug, 2015) 50. 10

    work page 2015

  15. [15]

    W. L. K. Wu, L. M. Mocanu, P. A. R. Ade, A. J. Anderson, J. E. Austermann, J. S. Avva, J. A. Beall, A. N. Bender, B. A. Bensonet al., Astrophys. J.884(Oct., 2019) 70,arXiv:1905.05777

    work page arXiv 2019

  16. [16]

    Millea, C

    M. Millea, C. M. Daley, T.-L. Chou, E. Anderes, P. A. R. Ade, A. J. Anderson, J. E. Austermann, J. S. Avva, J. A. Beall et al., Astrophys. J.922(Dec., 2021) 259,arXiv:2012.01709

    work page arXiv 2021

  17. [17]

    Z. Pan, F. Bianchini, W. L. K. Wu, P. A. R. Ade, Z. Ahmed, E. Anderes, A. J. Anderson, B. Ansarinejad, M. Archipley et al., Phys. Rev. D108(Dec., 2023) 122005,arXiv:2308.11608

    work page arXiv 2023

  18. [18]

    A 2500 square-degree CMB lensing map from combined South Pole Telescope and Planck data

    Y. Omori, R. Chown, G. Simard, K. T. Story, K. Aylor, E. J. Baxter, B. A. Benson, L. E. Bleem, J. E. Carlstromet al., Astrophys. J.849(2017), no. 2 124,arXiv:1705.00743

    work page internal anchor Pith review Pith/arXiv arXiv 2017

  19. [19]

    F. J. Quet al.,arXiv:2504.20038

    work page arXiv

  20. [20]

    P. A. R. Adeet al., Astron. Astrophys.594(2016) A14,arXiv:1502.01590. [24]PlanckCollaborationAstron. Astrophys.(2018)arXiv:1807.06209

    work page internal anchor Pith review Pith/arXiv arXiv 2016

  21. [21]

    M. S. Madhavacherilet al., Astrophys. J.962(2024), no. 2 113,arXiv:2304.05203

    work page internal anchor Pith review Pith/arXiv arXiv 2024

  22. [22]

    M. S. MadhavacherilPhil. Trans. Roy. Soc. Lond. A383(2025), no. 2290 20240025,arXiv:2411.08152

    work page arXiv 2025

  23. [23]

    Chudaykin, M

    A. Chudaykin, M. Kunz, and J. CarronPhys. Rev. D112(2025), no. 8 083537,arXiv:2503.09893

    work page arXiv 2025

  24. [24]

    Simons Observatory CollaborationJ. Cosmol. Astropart. Phys.02(2019) 056,arXiv:1808.07445

    work page internal anchor Pith review Pith/arXiv arXiv 2019

  25. [25]

    Simons Observatory CollaborationJ. Cosmol. Astropart. Phys.08(2025) 034,arXiv:2503.00636

    work page arXiv 2025

  26. [26]

    J. Liu, Z. Sun, J. Han, J. Carron, J. Delabrouille, S. Li, Y. Liu, J. Jin, S. Ghoshet al., Sci. China Phys. Mech. Astron. 65(2022), no. 10 109511,arXiv:2204.08158

    work page arXiv 2022

  27. [27]

    LiteBIRD CollaborationProg. Theor. Exp. Phys.2023(2023) 042F01,arXiv:2202.02773

    work page arXiv 2023

  28. [28]

    A. I. Lonappanet al., J. Cosmol. Astropart. Phys.06(2024) 009,arXiv:2312.05184

    work page arXiv 2024

  29. [29]

    Ruiz-Grandaet al., J

    M. Ruiz-Grandaet al., J. Cosmol. Astropart. Phys.11(2025) 073,arXiv:2507.22618

    work page arXiv 2025

  30. [30]

    Geet al., Phys

    F. Geet al., Phys. Rev. D111(2025), no. 8 083534,arXiv:2411.06000

    work page arXiv 2025

  31. [31]

    Jhaveri, T

    T. Jhaveri, T. Karwal, and W. HuarXiv:2504.21813

    work page arXiv

  32. [32]

    Mapping the Dark Matter through the CMB Damping Tail

    W. HuAstrophys. J.557(2001) L79–L83,astro-ph/0105424

    work page internal anchor Pith review Pith/arXiv arXiv 2001

  33. [33]

    CMB Lensing Reconstruction on the Full Sky

    T. Okamoto and W. HuPhys. Rev. D67(2003) 083002,astro-ph/0301031

    work page internal anchor Pith review Pith/arXiv arXiv 2003

  34. [34]

    Bias-Hardened CMB Lensing

    T. Namikawa, D. Hanson, and R. TakahashiMon. Not. R. Astron. Soc.431(2013) 609–620,arXiv:1209.0091. [39]PlanckCollaboration, N. Aghanim, Y. Akrami, M. Ashdown, J. Aumont, C. Baccigalupi, M. Ballardini, A. J. Banday, R. B. Barreiroet al., Astron. Astrophys.641(Sept., 2020) A8,arXiv:1807.06210

    work page internal anchor Pith review Pith/arXiv arXiv 2013

  35. [35]

    Kahn and A

    H. Kahn and A. W. MarshallOper. Res.1(1953), no. 5 263–278

    work page 1953

  36. [36]

    S. S. Lavenberg and P. D. WelchManagement Science27(1981), no. 3 322–335

    work page 1981

  37. [37]

    Chartier, B

    N. Chartier, B. Wandelt, Y. Akrami, and F. Villaescusa-NavarroMon. Not. R. Astron. Soc.503(2021), no. 2 1897–1914,arXiv:2009.08970

    work page arXiv 2021

  38. [38]

    Kokron, S.-F

    N. Kokron, S.-F. Chen, M. White, J. DeRose, and M. MausJ. Cosmol. Astropart. Phys.09(2022) 059, arXiv:2205.15327

    work page arXiv 2022

  39. [39]

    Ding, C.-H

    Z. Ding, C.-H. Chuang, Y. Yu, L. H. Garrison, A. E. Bayer, Y. Feng, C. Modi, D. J. Eisenstein, M. Whiteet al., Mon. Not. R. Astron. Soc.514(2022), no. 3 3308–3326,arXiv:2202.06074

    work page arXiv 2022

  40. [40]

    Chartier and B

    N. Chartier and B. D. WandeltMon. Not. R. Astron. Soc.515(2022), no. 1 1296–1315,arXiv:2204.03070

    work page arXiv 2022

  41. [41]

    DeRose, S.-F

    J. DeRose, S.-F. Chen, N. Kokron, and M. WhiteJ. Cosmol. Astropart. Phys.2023(2023), no. 02 008, arXiv:2210.14239

    work page arXiv 2023

  42. [42]

    Hadzhiyska, M

    B. Hadzhiyska, M. J. White, X. Chen, L. H. Garrison, J. DeRose, N. Padmanabhan, C. Garcia-Quintero, J. Mena-Fern´ andez, S.-F. Chenet al., The Open Journal of Astrophysics6(2023) 23,arXiv:2308.12343

    work page arXiv 2023

  43. [43]

    Doytcheva, F

    A. Doytcheva, F. V. Gerou, and J. U. LangearXiv e-prints(2024) arXiv:2410.14546,arXiv:2410.14546

    work page arXiv 2024

  44. [44]

    Hadzhiyska, R

    B. Hadzhiyska, R. de Belsunce, A. Cuceu, J. Guy, M. M. Ivanov, H. Coquinot, and A. Font-RiberaMon. Not. R. Astron. Soc.540(2025), no. 2 1960–1978,arXiv:2503.13442

    work page arXiv 2025

  45. [45]

    Kokron and S.-F

    N. Kokron and S.-F. ChenarXiv e-prints(2025) arXiv:2510.07375,arXiv:2510.07375

    work page arXiv 2025

  46. [46]

    Lensing reconstruction from PLANCK sky maps: inhomogeneous noise

    D. Hanson, G. Rocha, and K. GorskiMon. Not. R. Astron. Soc.400(2009) 2169–2173,arXiv:0907.1927

    work page internal anchor Pith review Pith/arXiv arXiv 2009

  47. [47]

    M. S. Madhavacheril, K. M. Smith, B. D. Sherwin, and S. NaessJ. Cosmol. Astropart. Phys.2021(11, 2021) 028, arXiv:2011.02475

    work page arXiv 2021

  48. [48]

    Weak lensing of the CMB

    D. Hanson, A. Challinor, and A. LewisGen. Rel. Grav.42(2010) 2197–2218,arXiv:0911.0612

    work page internal anchor Pith review Pith/arXiv arXiv 2010

  49. [49]

    Bianchini and A

    F. Bianchini and A. S. Maniyar,The Encyclopedia of Astrophysics, ch. The Cosmic Microwave Background – Secondary Anisotropies. Elsevier, 1, 2025.arXiv:2501.13913

    work page arXiv 2025

  50. [50]

    The shape of the CMB lensing bispectrum

    A. Lewis, A. Challinor, and D. HansonJ. Cosmol. Astropart. Phys.03(2011) 018,arXiv:1101.2234

    work page internal anchor Pith review Pith/arXiv arXiv 2011

  51. [51]

    CMB temperature lensing power reconstruction

    D. Hanson, A. Challinor, G. Efstathiou, and P. BielewiczPhys. Rev. D83(2011) 043005,arXiv:1008.4403

    work page internal anchor Pith review Pith/arXiv arXiv 2011

  52. [52]

    PICO: Probe of Inflation and Cosmic Origins

    S. Hananyet al.,arXiv:1902.10541

    work page internal anchor Pith review Pith/arXiv arXiv 1902

  53. [53]

    Analysis of CMB polarization on an incomplete sky

    A. Lewis, A. Challinor, and N. TurokPhys. Rev. D65(2002) 023505,astro-ph/0106536

    work page internal anchor Pith review Pith/arXiv arXiv 2002

  54. [54]

    E. F. Bunn, M. Zaldarriaga, M. Tegmark, and A. de Oliveira-CostaPhys. Rev. D67(2003) 023501,astro-ph/0207338

    work page internal anchor Pith review Pith/arXiv arXiv 2003

  55. [55]

    K. M. SmithPhys. Rev. D74(2006) 083002,astro-ph/0511629

    work page internal anchor Pith review Pith/arXiv arXiv 2006

  56. [56]

    Namikawaet al., Phys

    T. Namikawaet al., Phys. Rev. D101(2020), no. 8 083527,arXiv:2001.10465

    work page arXiv 2020

  57. [57]

    Constraints on Patchy Reionization from Planck CMB Temperature Trispectrum

    T. NamikawaPhys. Rev. D97(2018), no. 6 063505,arXiv:1711.00058

    work page internal anchor Pith review Pith/arXiv arXiv 2018

  58. [58]

    Full-sky lensing reconstruction of gradient and curl modes from CMB maps

    T. Namikawa, D. Yamauchi, and A. TaruyaJ. Cosmol. Astropart. Phys.1201(2012) 007,arXiv:1110.1718

    work page internal anchor Pith review Pith/arXiv arXiv 2012

  59. [59]

    Mass Reconstruction with CMB Polarization

    W. Hu and T. OkamotoAstrophys. J.574(2002) 566–574,astro-ph/0111606

    work page internal anchor Pith review Pith/arXiv arXiv 2002