dénes sexty - ippp conference management system (indico)...zoltán fodor, sándor katz, csaba...
TRANSCRIPT
![Page 1: Dénes Sexty - IPPP Conference Management System (Indico)...Zoltán Fodor, Sándor Katz, Csaba Török . Proof of convergence for CLE results If there is fast decay and a holomorphic](https://reader033.vdocument.in/reader033/viewer/2022041620/5e3ea9f7813e682cdc69d0d3/html5/thumbnails/1.jpg)
The issue of the poles for lattice models
Comparison of CLE and reweighting for QCD
Dénes SextyWuppertal University, Eötvös University
Lattice 2016, Southampton, 26th of July, 2016
1. poles and HDQCD2. poles and full QCD [Aarts, Seiler, Sexty, Stamatescu in prep.]3. Comparison of CLE and reweighting for full QCD [Fodor, Katz, Sexty, Torok (2015)]
Collaborators: Gert Aarts, Erhard Seiler, Ion-Olimpiu Stamatescu Felipe Attanasio, Lorenzo Bongiovanni, Benjamin Jäger, Zoltán Fodor, Sándor Katz, Csaba Török
![Page 2: Dénes Sexty - IPPP Conference Management System (Indico)...Zoltán Fodor, Sándor Katz, Csaba Török . Proof of convergence for CLE results If there is fast decay and a holomorphic](https://reader033.vdocument.in/reader033/viewer/2022041620/5e3ea9f7813e682cdc69d0d3/html5/thumbnails/2.jpg)
Proof of convergence for CLE results
If there is fast decay
and a holomorphic action
[Aarts, Seiler, Stamatescu (2009) Aarts, James, Seiler, Stamatescu (2011)]
then CLE converges to the correct result
P (x , y )→0 as x , y→∞
S (x)
S=SW [U μ]+ln DetM (μ) measure has zeroscomplex logarithm has a branch cut meromorphic drift
Non-holomorphic action for nonzero density(Det M=0)
[Mollgaard, Splittorff (2013), Greensite(2014)]
Incorporating poles to proof, investigations of toy models
[See Gert Aarts' talk]
![Page 3: Dénes Sexty - IPPP Conference Management System (Indico)...Zoltán Fodor, Sándor Katz, Csaba Török . Proof of convergence for CLE results If there is fast decay and a holomorphic](https://reader033.vdocument.in/reader033/viewer/2022041620/5e3ea9f7813e682cdc69d0d3/html5/thumbnails/3.jpg)
Heavy Quark QCD at nonzero chemical potential (HDQCD)
DetM (μ)=∏xdet (1+C P x)
2 det (1+C ' P x−1)2
P x=∏τU 0( x+τa0) C=[2 κexp(μ)]N τ C '=[2κexp(−μ)]N τ
Hopping parameter expansion of the fermion determinantSpatial hoppings are dropped
S=SW [U μ]+ln DetM (μ)
Studied with reweighting [De Pietri, Feo, Seiler, Stamatescu (2007)][Rindlischbacher, de Forcrand (2015)]
CLE study using gaugecooling
[Seiler, Sexty, Stamatescu (2012)][Aarts, Attanasio, Jäger, Sexty (2016)]
R=e∑
xC Tr Px+C ' Tr P−1
![Page 4: Dénes Sexty - IPPP Conference Management System (Indico)...Zoltán Fodor, Sándor Katz, Csaba Török . Proof of convergence for CLE results If there is fast decay and a holomorphic](https://reader033.vdocument.in/reader033/viewer/2022041620/5e3ea9f7813e682cdc69d0d3/html5/thumbnails/4.jpg)
Critical chemical potential in HDQCD
DetM (μ)=∏xdet (1+2κ eμ P x)
2 det (1+2κ e−μ P x−1)2
⟨exp(2 iϕ)⟩= ⟨ DetM (μ)
DetM (−μ) ⟩Phase average
Hard sign problem 1<μ<1.8
Except in the middle at half filling μc=−ln(2κ)
det (1+C P )=1+C3+C Tr P+C 2 Tr P−1
At only the second factor has a(n exponentially suppressed) sign problemμc
![Page 5: Dénes Sexty - IPPP Conference Management System (Indico)...Zoltán Fodor, Sándor Katz, Csaba Török . Proof of convergence for CLE results If there is fast decay and a holomorphic](https://reader033.vdocument.in/reader033/viewer/2022041620/5e3ea9f7813e682cdc69d0d3/html5/thumbnails/5.jpg)
Do poles play a role in HDQCD?
Distribution around the zero of the determinant
Only gets close to the pole around μc
Otherwise the pole is outside of the distribution
Where it shows criticality
Worst case for poles: zero temperature lattice
![Page 6: Dénes Sexty - IPPP Conference Management System (Indico)...Zoltán Fodor, Sándor Katz, Csaba Török . Proof of convergence for CLE results If there is fast decay and a holomorphic](https://reader033.vdocument.in/reader033/viewer/2022041620/5e3ea9f7813e682cdc69d0d3/html5/thumbnails/6.jpg)
Distribution of the local determinants on the complex plane
μ=1.3
μ=1.425=μc
Distribution close to real axis, but “touches” pole
Very faint “whiskers”Similar to the toy model caseNegligible contribution to averages
Well separated from poles
Exact results
![Page 7: Dénes Sexty - IPPP Conference Management System (Indico)...Zoltán Fodor, Sándor Katz, Csaba Török . Proof of convergence for CLE results If there is fast decay and a holomorphic](https://reader033.vdocument.in/reader033/viewer/2022041620/5e3ea9f7813e682cdc69d0d3/html5/thumbnails/7.jpg)
Conclusion for HDQCD
Results are unaffected by poles almost everywhere
Near the critical chemical potential we have indications that results are probably OK affected by a negligibly small contamination
Phase diagram mapped out with complex Langevin
[Aarts, Attanasio, Jäger, Sexty arxiv:1606.05561]
[See Felipe Attanasio's talk]
![Page 8: Dénes Sexty - IPPP Conference Management System (Indico)...Zoltán Fodor, Sándor Katz, Csaba Török . Proof of convergence for CLE results If there is fast decay and a holomorphic](https://reader033.vdocument.in/reader033/viewer/2022041620/5e3ea9f7813e682cdc69d0d3/html5/thumbnails/8.jpg)
Full QCD and the issue of poles
[Sexty (2014), Aarts, Seiler, Sexty, Stamatescu (2015)]
Unimproved staggered and Wilson fermions with CLE
Seff=Sg(U )−N f ln det M (U )
=Sg(U )−N f ∑iln λi(U )
Drift term of fermions
K f=N f∑i
D λi(U )
λ i(U )
Poles can be an issue if eigenvalue density around zero is not vanishing
Total phase of the determinant is sum off all the phases
Sign problem can still be hard
![Page 9: Dénes Sexty - IPPP Conference Management System (Indico)...Zoltán Fodor, Sándor Katz, Csaba Török . Proof of convergence for CLE results If there is fast decay and a holomorphic](https://reader033.vdocument.in/reader033/viewer/2022041620/5e3ea9f7813e682cdc69d0d3/html5/thumbnails/9.jpg)
Spectrum of the Dirac operator above the deconfinement transition
![Page 10: Dénes Sexty - IPPP Conference Management System (Indico)...Zoltán Fodor, Sándor Katz, Csaba Török . Proof of convergence for CLE results If there is fast decay and a holomorphic](https://reader033.vdocument.in/reader033/viewer/2022041620/5e3ea9f7813e682cdc69d0d3/html5/thumbnails/10.jpg)
The phase of the determinant
Langevin time evolution
HistogramConclusions for full QCD
At high temperatureseigenvalue density is zero at the originEven tough the sign problem can be hard
At low temperaturesNon-zero eigenvalue density is expected (Banks-Casher relation)Can we deal with it?
![Page 11: Dénes Sexty - IPPP Conference Management System (Indico)...Zoltán Fodor, Sándor Katz, Csaba Török . Proof of convergence for CLE results If there is fast decay and a holomorphic](https://reader033.vdocument.in/reader033/viewer/2022041620/5e3ea9f7813e682cdc69d0d3/html5/thumbnails/11.jpg)
⟨F ⟩μ=∫DU e−S E det M (μ)F
∫DU e−S E det M (μ)=∫DU e−S E R
det M (μ)
RF
∫DU e−S E Rdet M (μ)
R
=⟨F det M (μ)/R ⟩R
⟨det M (μ)/R ⟩R
Reweighting
⟨ det M (μ)
R ⟩R
=Z (μ)
Z R=exp (−VT Δ f (μ , T ))
Δ f (μ , T ) =free energy difference
Exponentially small as the volume increases
Reweighting works for large temperatures and small volumes
⟨F ⟩μ → 0 /0
μ/T≈1Sign problem gets hard at
R=det M (μ=0), ∣det M (μ)∣, etc.
![Page 12: Dénes Sexty - IPPP Conference Management System (Indico)...Zoltán Fodor, Sándor Katz, Csaba Török . Proof of convergence for CLE results If there is fast decay and a holomorphic](https://reader033.vdocument.in/reader033/viewer/2022041620/5e3ea9f7813e682cdc69d0d3/html5/thumbnails/12.jpg)
Comparison with reweighting for full QCD
[Fodor, Katz, Sexty, Török 2015]
R=DetM (μ=0)
Reweighting from ensemble at
![Page 13: Dénes Sexty - IPPP Conference Management System (Indico)...Zoltán Fodor, Sándor Katz, Csaba Török . Proof of convergence for CLE results If there is fast decay and a holomorphic](https://reader033.vdocument.in/reader033/viewer/2022041620/5e3ea9f7813e682cdc69d0d3/html5/thumbnails/13.jpg)
Overlap problem
Histogram of weights Relative to the largest weight in ensemble
Average becomes dominated by very few configurations
![Page 14: Dénes Sexty - IPPP Conference Management System (Indico)...Zoltán Fodor, Sándor Katz, Csaba Török . Proof of convergence for CLE results If there is fast decay and a holomorphic](https://reader033.vdocument.in/reader033/viewer/2022041620/5e3ea9f7813e682cdc69d0d3/html5/thumbnails/14.jpg)
Sign problem
Sign problem gets hard around μ/T≈1−1.5
⟨exp(2 iϕ)⟩= ⟨det M (μ)
det M (−μ) ⟩
![Page 15: Dénes Sexty - IPPP Conference Management System (Indico)...Zoltán Fodor, Sándor Katz, Csaba Török . Proof of convergence for CLE results If there is fast decay and a holomorphic](https://reader033.vdocument.in/reader033/viewer/2022041620/5e3ea9f7813e682cdc69d0d3/html5/thumbnails/15.jpg)
Comparisons as a function of beta
at N T=4 breakdown at β=5.1 − 5.2
Similarly to HDQCD Cooling breaks down at small beta
At larger NT ?
![Page 16: Dénes Sexty - IPPP Conference Management System (Indico)...Zoltán Fodor, Sándor Katz, Csaba Török . Proof of convergence for CLE results If there is fast decay and a holomorphic](https://reader033.vdocument.in/reader033/viewer/2022041620/5e3ea9f7813e682cdc69d0d3/html5/thumbnails/16.jpg)
Comparisons as a function of beta
NT=8NT=6
Breakdown prevents simulations in the confined phase
for staggered fermions with N T=4,6,8
mπ≈2.3T c
Two ensembles: mπ≈4.8 T c
![Page 17: Dénes Sexty - IPPP Conference Management System (Indico)...Zoltán Fodor, Sándor Katz, Csaba Török . Proof of convergence for CLE results If there is fast decay and a holomorphic](https://reader033.vdocument.in/reader033/viewer/2022041620/5e3ea9f7813e682cdc69d0d3/html5/thumbnails/17.jpg)
Conclusions
Zeroes of the measure can affect validity of CLE if prob. density around them is non-vanising In HDQCD poles only have a negligible effect around critical chemical potential, otherwise exactIn full QCD high temperature simulations are OK Low temperatures?
Comparison of reweighting with CLE they agree where both works Reliability can be judged independent of the other method Low temperature phase not yet reached with
NT=8