Fault-tolerant quantum computing with color codes Andrew J. Landahl with Jonas T. Anderson and Patrick R. Rice. arxiv:

Transcription

1 '!" '!"!!!* % $!!!* "#$%!"#$#%&'()*+,&-.&"#/-0#.$)(&'#%-1#.&& & 2-$"&+3+.&)4#5.$&#6&2"-$+&%')*+&!! &7+$2++.&'"#$#%&).,&"+),+/& "#$% &!&!!!%!$ % $ &+!! ) '!# )!+' ( %+!&!" (!& *+!" )+!% (+!$!* % $!!!%!$ #+!* % $ $+!# ) ' (!"!#!*!&!# ) (!&!#!%!$!%!$ Fault-tolerant quantum computing with color codes Andrew J. Landahl with Jonas T. Anderson and Patrick R. Rice. arxiv: :;:88& This work was supported in part by the Laboratory Directed Research and Development program at Sandia National Laboratories. Sandia National Laboratories is a multi-program laboratory managed and operated by Sandia Corporation, a wholly owned subsidiary of Lockheed Martin Corporation, for the U.S. Department of Energy s National Nuclear Security Administration under contract DE-AC04-94AL <=>88?&@+*#.,&A.$+/.)1#.)(&>#.6+/+.*+&#.&<5).$54&=//#/&>#//+*1#.&

2 Photos placed in horizontal posi1on with even amount of white space between photos and header Fault-tolerant quantum computing with color codes Andrew J. Landahl with Jonas T. Anderson and Patrick R. Rice. arxiv: /8/11 This work was supported in part by the Laboratory Directed Research and Development program at Sandia National Laboratories. Sandia National Laboratories is a multi-program laboratory managed and operated by Sandia Corporation, a wholly owned subsidiary of Lockheed Martin Corporation, for the U.S. Department of Energy s National Nuclear Security Administration under contract DE-AC04-94AL QEC11: Second Interna1onal Conference on Quantum Error Correc1on

3 Color codes The three semiregular 2D topological color codes The 2D topological subsystem color code [Bombin & Mar1n Delgado, PRL 97, (2006)] [Bombin, PRA 80, (2010)] S is transversal Fewest qubits/distance Planar color codes: 3m corners Checks: X f,z f S is transversal Two body checks suffice These codes are naturally suited to 2D quantum technologies in which long distance quantum transport is imprac1cal. The 2D surface code has many promising features for fault tolerant quantum compu1ng, including a high accuracy threshold and no need for syndrome ancilla dis4lla4on. How do 2D color codes compare? QEC11: Second Interna1onal Conference on Quantum Error Correc1on 3

4 Control & noise models Control model: (Faulty) gate basis: {I, X, Z, H, S, S, CNOT,M Z,M X, 0, +, π/4 } Standard assumpdons: Parallel opera1on, refreshable ancilla, fast classical computa1on, equal 1me gates. Locality assumpdons: 2D layout, local quantum processing. BP channel: Bit flip channel B(p) followed by phase flip channel Φ(p). DP channel: Applies each two qubit ( double Pauli ) product with probability p/16. Noise model: Standard assumpdons: No leakage, reliable classical computa1on. 1. Circuit based noise model Each prepara1on and one qubit gate followed by BP(p). Each CNOT gate followed by DP(p). Each measurement preceded by BP(p) and result flipped with probability p. 2. Phenomenological noise model Same, except each syndrome bit extrac1on circuit modeled phenomenologically as a measurement that fails with probability p; ignores noise propaga1on between data and ancilla. Gates only appear in encoded computa1on. 3. Code capacity noise model Same as phenomenological model, except syndrome measurements are perfect. QEC11: Second Interna1onal Conference on Quantum Error Correc1on 4

5 Decoders & thresholds OpDmal decoder: Returns recovery most likely to succeed given the syndrome. MLE decoder: Returns most likely error that occurred given the syndrome. 7.8% [A] 0.9% [B] [A: Sarvepalli & Raussendorf, arxiv: ] [B: Fowler, Whiteside, and Hollenberg, arxiv: ] [See our paper for other references.] QEC11: Second Interna1onal Conference on Quantum Error Correc1on 5

6 Syndrome extraction XZ sequendal schedule XZ interleaved schedule 0 M Z + M X Example of error propagadon X error on X check bit (red circle) between 1me steps 5 and 6. Propagates to 3 data errors; detected correctly by 3 Z check bits (yellow) QEC11: Second Interna1onal Conference on Quantum Error Correc1on 6

7 Decoding Code capacity MLE decoder: (Works for all CSS codes.) OpDmizaDon problem Integer program over GF(2) Integer program over the reals min v x v sto x v = s f f v f x v B := {0, 1} min 1 T x sto Hx = s mod 2 x B n min c T y sto Ay = z y B n z := s +2t 1 +4t 2 +8t 3 y := (x T, t T 1, t T 2, t T 3 ) T c := ( 1 T, 0 T, 0 T, 0 T ) T A := (H 2I 4I 8I) Phenomenological MLE decoder: (Works for all CSS codes.) Integer program over the reals Measurement error Data error min c T y sto Ay = z y B n H H A :=... H I I z := s +2t 1 +4t 2 +8t 3 y := (x T data, x T synd, t T 1, t T 2, t T 3 ) T c := ( 1 T, 1 T, 0 T, 0 T, 0 T ) T I. 2I 4I 8I... I I QEC11: Second Interna1onal Conference on Quantum Error Correc1on 7

8 Code-capacity threshold Exact curves found up to d = 7. p fail = p E (1 p) n E, failing patterns E p (est) fail Monte Carlo esdmate for d = 9. = N fail N (σ 2 fail) (est) = p (est) fail ( 1 p (est) fail N ) Threshold by scaling Ansatz fit, not curve crossing esdmate. [Wang, Harrington, & Preskill, Ann. Phys. 303, 31 (2003)] Theory: Fit: ξ p p c ν 0 p fail =(p p c )d 1/ν 0 p fail = A + B(p p c )d 1/ν 0 p th = 10.56(1)% N.B. Finite size effects may ma:er. P Failure (a) P Error (b) d=3 d=5 d=7 d= QEC11: Second Interna1onal Conference on Quantum Error Correc1on 8

9 Phenomenological threshold Algorithm 1 : p fail (p) by Monte Carlo 1: n faces 1 4 (d + 1)2 1. 2: for i =1to N do d=5 d=7 d=9 3: // Generate data and syndrome errors for d time slices. 4: for t =1to d do 5: for j =1to n do 6: E[t, j] 1 with probability p. // Data errors. 7: end for 8: for j = n +1to n n faces do 9: E[t, j] 1 with probability p. // Synd. errors. 10: end for 11: end for 12: E min Decode(Syndrome(E)). // 3D error volume. 13: E L t E[t] E min[t]. // 2D error plane. 14: E min Decode(Syndrome(E )). // Ideal decoding. 15: if ( L i E [i] E min[i] = 1) then 16: N fail N fail : end if 18: end for 19: return p (est) fail = N fail /N. P Failure P Failure P Error d=5 d=7 d=9 p th =3.05(4)% P Error QEC11: Second Interna1onal Conference on Quantum Error Correc1on 9

10 Circuit threshold d=5 d=7 d= d=5 d=7 d= P Failure P Failure P Error P Error Circuit generates correlated errors up to weight four ( hooks ). Can accommodate by modifying fi^ng funcdon: p fail = A + B(p p c )d 1/ν 0 + C(p p c ) 2 d 2/ν 0. (XZ interleaved) p th (XZ sequential) p th =8.2(3) 10 4 =8.0(4) 10 4 QEC11: Second Interna1onal Conference on Quantum Error Correc1on 10

11 p fail = p(o E + E min ) = 2 O log ops O string-net O SAW( d) Self-avoiding walk bound p(o E + E min )+ p(o E + E min ) p(o E + E min ) O string 2 n n SAW (L) 2 L [p(1 p)] L/2 L d # star1ng points for SAW # ways to choose flips on E + E min Prob. flips at those loca1ons n SAW (L) max ( max 1) L 1 n SAW (L) P (L) µ L p th (1 p th ) 1 4µ 2 Code capacity noise model µ (6) µ p th (1)% cf (1)% Phenomenological noise model max = 10 p th % cf. 3.05(4)% [Dennis, Kitaev, Landahl, & Preskill, JMP 43, 4452 (2002)] QEC11: Second Interna1onal Conference on Quantum Error Correc1on 11

12 Architectures & computation Transversal Push noise channels through gates to get effec1ve noise channels for FTQEC Magic states for T gate injected by teleporta1on. 0 M M X / H Unencode M Z 0 / X Z M p I = p X = p Z = p 0 = p + = 3 2 p CNOT = 3 2 p(nondestructive) M X =2p H =2p (bit-flip) S =8.2(3) 10 4 p (destructive) M X =2p (bit-flip) T = p (destructive) M Z = 10.56(1)% p π/4 =( 2 1)/ % =3p (phase-flip) S =3p (bit-flip) T = 3 2 p(nondestructive) M Z Code deformadon Local ops to grow, shrink, & move defects. H, S, CNOT gate by code deforma1on. Magic states for T gate injected by code deforma1on. QEC11: Second Interna1onal Conference on Quantum Error Correc1on 12

13 Summary Color codes vs. surface codes Code capacity & phenomenological thresholds comparable to surface codes. (11%, 3%) Circuit model threshold about 10 1mes smaller than surface code s. Rigorous bounds on threshold very weak at this point. Qubit overhead comparable to surface codes. Open quesdons: TSCCs: Do they have beper thresholds? Leakage errors: How well are they tolerated? Efficient decoders: TCC = 2 surface codes run efficient matching on these. Threshold? Rigorous lower bounds: Tighter analysis techniques? Magic state injec1on: How much of 14.6% threshold is consumed by imperfect injec1on? Topological quantum computa1on: How to formulate a color code quantum double model? QEC11: Second Interna1onal Conference on Quantum Error Correc1on 13