1 / 153

Adaptive annealing: a near-optimal connection between sampling and counting

Adaptive annealing: a near-optimal connection between sampling and counting. Daniel Štefankovič (University of Rochester) Santosh Vempala Eric Vigoda (Georgia Tech). Adaptive annealing: a near-optimal connection between sampling and counting. If you want to count

dlopez
Download Presentation

Adaptive annealing: a near-optimal connection between sampling and counting

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Adaptive annealing: a near-optimal connectionbetween sampling and counting Daniel Štefankovič (University of Rochester) Santosh Vempala Eric Vigoda (Georgia Tech)

  2. Adaptive annealing: a near-optimal connectionbetween sampling and counting If you want to count using MCMC then statistical physics is useful. Daniel Štefankovič (University of Rochester) Santosh Vempala Eric Vigoda (Georgia Tech)

  3. Outline 1. Counting problems 2. Basic tools: Chernoff, Chebyshev 3. Dealing with large quantities (the product method) 4. Statistical physics 5. Cooling schedules (our work) 6. More…

  4. Counting independent sets spanning trees matchings perfect matchings k-colorings

  5. Counting independent sets spanning trees matchings perfect matchings k-colorings

  6. Compute the number of spanning trees

  7. Compute the number of spanning trees det(D – A)vv Kirchhoff’s Matrix Tree Theorem: - det D A

  8. Compute the number of spanning trees polynomial-time algorithm G number of spanning trees of G

  9. ? Counting independent sets spanning trees matchings perfect matchings k-colorings

  10. Compute the number of independent set subset S of vertices, of a graph no two in S are neighbors = independent sets (hard-core gas model)

  11. # independent sets = 7 independent set = subset S of vertices no two in S are neighbors

  12. # independent sets = G1 G2 G3 ... Gn-2 ... Gn-1 ... Gn

  13. # independent sets = 2 G1 3 G2 5 G3 ... Gn-2 Fn-1 ... Gn-1 Fn ... Gn Fn+1

  14. # independent sets = 5598861 independent set = subset S of vertices no two in S are neighbors

  15. Compute the number of independent sets ? polynomial-time algorithm G number of independent sets of G

  16. Compute the number of independent sets (unlikely) ! polynomial-time algorithm G number of independent sets of G

  17. graph G  # independent sets in G #P NP FP P #P-complete #P-complete even for 3-regular graphs (Dyer, Greenhill, 1997)

  18. graph G  # independent sets in G ? approximation randomization

  19. graph G  # independent sets in G ? which is more important? approximation randomization

  20. graph G  # independent sets in G My world-view: (true) randomness is important conceptually but NOT computationally (i.e., I believe P=BPP). approximation makes problems easier (i.e., I believe #P=BPP) ? which is more important? approximation randomization

  21. We would like to know Q Goal: random variable Y such that P( (1-)Q  Y  (1+)Q )  1- “Y gives (1)-estimate”

  22. We would like to know Q Goal: random variable Y such that P( (1-)Q  Y  (1+)Q )  1- (fully polynomial randomized approximation scheme): FPRAS: Y polynomial-time algorithm G,,

  23. Outline 1. Counting problems 2. Basic tools: Chernoff, Chebyshev 3. Dealing with large quantities (the product method) 4. Statistical physics 5. Cooling schedules (our work) 6. More...

  24. We would like to know Q X1 + X2 + ... + Xn Y= n 1. Get an unbiased estimator X, i. e., E[X] = Q 2. “Boost the quality” of X:

  25. The Bienaymé-Chebyshev inequality P( Y gives (1)-estimate ) 1 V[Y] 1 - E[Y]2 2

  26. The Bienaymé-Chebyshev inequality P( Y gives (1)-estimate ) 1 V[Y] 1 - E[Y]2 2 squared coefficient of variation SCV X1 + X2 + ... + Xn V[Y] V[X] 1  Y= = n E[Y]2 E[X]2 n

  27. The Bienaymé-Chebyshev inequality X1 + X2 + ... + Xn Y= n Let X1,...,Xn,X be independent, identically distributed random variables, Q=E[X]. Let Then P( Y gives (1)-estimate of Q ) 1 V[X] 1 - 2 n E[X]2

  28. Chernoff’s bound X1 + X2 + ... + Xn Y= n Let X1,...,Xn,X be independent, identically distributed random variables, 0  X  1, Q=E[X]. Let Then P( Y gives (1)-estimate of Q ) - 2 . n . E[X] / 3 e  1 –

  29. 1 1 V[X] n = 2  E[X]2 Number of samples to achieve precision  with confidence . 3 1 ln (1/) n = 2 E[X] 0X1

  30. BAD 1 1 V[X] n = 2  E[X]2 Number of samples to achieve precision  with confidence . 3 1 ln (1/) n = 2 E[X] GOOD 0X1 BAD

  31. Median “boosting trick” X1 + X2 + ... + Xn Y= n 1 4 n = 2 E[X] BY BIENAYME-CHEBYSHEV:  )  3/4 P( (1-)Q (1+)Q = Y

  32. Median trick – repeat 2T times (1-)Q (1+)Q BY BIENAYME-CHEBYSHEV:  )  3/4 P(  BY CHERNOFF: -T/4 > T out of 2T )  1 - e P(  -T/4 median is in )  1 - e P(

  33. V[X] 32 n = ln (1/) 2 E[X]2 + median trick 3 1 n = ln (1/) 2 E[X] 0X1 BAD

  34. ( ) 1 V[X] ln (1/) n = 2 E[X]2 Creating “approximator” from X  = precision  = confidence

  35. Outline 1. Counting problems 2. Basic tools: Chernoff, Chebyshev 3. Dealing with large quantities (the product method) 4. Statistical physics 5. Cooling schedules (our work) 6. More...

  36. (approx) counting  sampling Valleau,Card’72 (physical chemistry), Babai’79 (for matchings and colorings), Jerrum,Valiant,V.Vazirani’86 the outcome of the JVV reduction: random variables: X1 X2 ... Xt such that E[X1 X2 ... Xt] 1) = “WANTED” 2) the Xi are easy to estimate V[Xi] squared coefficient of variation (SCV) = O(1) E[Xi]2

  37. (approx) counting  sampling E[X1 X2 ... Xt] 1) = “WANTED” 2) the Xi are easy to estimate V[Xi] = O(1) E[Xi]2 Theorem (Dyer-Frieze’91) O(t2/2) samples (O(t/2) from each Xi) give 1 estimator of “WANTED” with prob3/4

  38. JVV for independent sets GOAL: given a graph G, estimate the number of independent sets of G 1 # independent sets = P( )

  39. P(AB)=P(A)P(B|A) JVV for independent sets ? ? ? ? ? ? P() = P() P() P( ) P( ) X1 X2 X3 X4 V[Xi] Xi [0,1] and E[Xi] ½  = O(1) E[Xi]2

  40. P(AB)=P(A)P(B|A) JVV for independent sets ? ? ? ? ? ? P() = P() P() P( ) P( ) X1 X2 X3 X4 V[Xi] Xi [0,1] and E[Xi] ½  = O(1) E[Xi]2

  41. Self-reducibility for independent sets P( ) ? 5 = ? 7 ?

  42. Self-reducibility for independent sets P( ) ? 5 = ? 7 ? 7 = 5

  43. Self-reducibility for independent sets P( ) ? 5 = ? 7 ? 7 7 = = 5 5

  44. Self-reducibility for independent sets P( ) 3 = ? 5 ? 5 = 3

  45. Self-reducibility for independent sets P( ) 3 = ? 5 ? 5 5 = = 3 3

  46. Self-reducibility for independent sets 7 5 7 = = 5 3 5 7 5 3 = 7 = 5 3 2

  47. JVV: If we have a sampler oracle: random independent set of G SAMPLER ORACLE graph G then FPRAS using O(n2) samples.

  48. JVV: If we have a sampler oracle: random independent set of G SAMPLER ORACLE graph G then FPRAS using O(n2) samples. ŠVV: If we have a sampler oracle: SAMPLER ORACLE set from gas-model Gibbs at  , graph G then FPRAS using O*(n) samples.

  49. Application – independent sets O*( |V| ) samples suffice for counting Cost per sample (Vigoda’01,Dyer-Greenhill’01) time = O*( |V| ) for graphs of degree  4. Total running time: O* ( |V|2 ).

More Related