N. Bansal 1, M. Charikar 2, R. Krishnaswamy 2, S. Li 3 1 TU Eindhoven 2 Princeton University 3 TTIC Midwest Theory Day, Purdue, May 3, 2014
Introduction Discrepancy Problem Broadcast Scheduling Problem Our Results and Techniques Negative Results O(log 1.5 n)-Approximation Outline
input ground set U a family S of subsets of U output: coloring minimize worst discrepancy: Discrepancy Problem U : {1,2,3,4,5,6} S : χ : χ : {1,3,5,6} {2,3,4,6} {1,4,5,6} {1,3,5,6} 0 {2,3,4,6} 0 {1,4,5,6} 0 {1,3,5,6} 0 {2,3,4,6} 0 {1,4,5,6} 2
S contains all subsets discrepancy = n/2 -disc. by randomized coloring -disc. (non-constructive) [Spencer 85] -disc. (constructive) [Bansal 10] [Lovett- Meka 12] -lower bound Interesting When |U| = | S | = n
Erdos’s Discrepancy U = {0, 1, 2, 3,......} S = {all arithmetic progressions starting at 0} open problem: is discrepancy bounded? Rectangle Discrepancy U = {n points in a 2-D plane} S = {all axis-parallel rectangles} discrepancy? ( between Ω(log n) and O(log 2.5 n) ) Special Discrepancy Problems
give 3 permutations of [n] find a coloring χ : [n] {±1} minimize the maximum discrepancy over all prefixes of the permutations 3-Permutation Discrepancy χ : discrepancy = 2
1 permutation : discrepancy=1, trivial 2 permutations : discrepancy=1, easy exercise 3 permutations? upper bound : O(log n) lower bound [Newman-Nikolov 11]: Ω(log n) l ≥ 3 permutations upper bound : O(l 1/2 log n) lower bound : max{Ω(l 1/2 ), Ω(log n)} Why 3 Permutations?
Introduction Discrepancy Problem Broadcast Scheduling Problem Our Results and Techniques Negative Results O(log 1.5 n)-Approximation Outline
a server holding n pages requests come over time broadcast 1 page per time slot minimize average response time offline version Broadcast Scheduling Problem response time = 2 Time response time =
Resource Allocation Scheduling Theory
NP-hard [Erlebach-Hall] (1/α)-speed,1/(1-2α)-approximation, α ≤ 1/3 [Kalyanasundaram et al.] (1/α)-speed: broadcast a page only requires α time slots (1+ε)-speed, O(1/ε) approximation, ε > 0[Bansal- Charikar-Khanna-Naor 05] O(n)-approx: trivial, cyclic order O(n 1/2 )-approx [Bansal-Charikar-Khanna-Naor 05] O(log 2 n)-approx[Bansal-Coppersmith-Sviridenko 08] Known Results
Introduction Discrepancy Problem Broadcast Scheduling Problem Our Results and Techniques Negative Results O(log 1.5 n)-Approximation Outline
previous bestour results approximationO(log 2 n)O(log 3/2 n) integrality gap1 + tiny constΩ(log n) hardnessNP-hardΩ(log 1/2 n) Our Results and Techniques negative results (integrality gap and hardness) connection to permutation discrepancy positive result Lovett-Meka algorithmic framework for discrepancy minimization
Introduction Discrepancy Problem Broadcast Scheduling Problem Our Results and Techniques Negative Results O(log 1.5 n)-Approximation Outline
Main Lemma Negative Results l-permutation instance Π broadcast scheduling instance I = = “discrepancy” optimal response time LP(I) = O(1) Main + Ω(log n)-disc. for 3-perm. Ω(log n)-int. gap Main + Ω(l 1/2 )-hard. for l-perm.(new) Ω(log 1/2 n)-hard.
Fractional Schedule from LP integral schedule fractional schedule Time response time 0.4 × × 2=1.6 requests
Main Lemma l-permutation instance Π broadcast scheduling instance I = = “discrepancy” optimal response time LP(I) = O(1) proof steps: construction of BS instance from l permutations Θ(1) LP value small discrepancy small response time small response time small discrepancy
given 3 permutations π 1 π 2 π 3 of size m π 1 = (5, 8, 4, 6, 3, 2, 1, 7) π 2 = (6, 7, 3, 8, 5, 1, 2, 4) π 3 = (7, 1, 3, 2, 8, 5, 6, 4) Construction of BS Instance π1π1 π2π2 π3π3 forbidden interval P1P1 P1P1 P2P2 P2P2 P3P3 P3P3 P4P4 P4P4 P5P5 P5P5 P6P6 P6P6 P7P7 P7P7 permutation interval Req: m/2
average response time ≈ # bad requests new goal: minimize #bad requests a request in P i is good if it is satisfied at P i or P i+1 otherwise, the request is bad Good and Bad Requests P1P1 P1P1 P2P2 P2P2 P3P3 P3P3 P4P4 P4P4 P5P5 P5P5 P6P6 P6P6 P7P7 P7P Req: Brd:
Req: LP solution each time slot, broadcast ½ fraction of each page requested P 7 : broadcast ½ fraction of the m pages arbitrarily all requests are good: ½ of request in P i is satisfied immediately remaining ½ satisfied at P i+1 Θ(1) LP Value P1P1 P1P1 P2P2 P2P2 P3P3 P3P3 P4P4 P4P4 P5P5 P5P5 P6P6 P6P6 P7P7 P7P7 request ½ satisfied
How to Make All Requests Good in an Integral Schedule? P1P1 P1P1 P2P2 P2P2 P3P3 P3P3 P4P4 P4P4 P5P5 P5P5 P6P6 P6P6 P7P7 P7P7 all m pages requested in all intervals(except P 7 ) each P-interval has m/2 slots solution: m/2 pages are broadcast in P 1, P 3, P 5, P 7 m/2 pages are broadcast in P 2, P 4, P 6 giving a balanced ±1 coloring of the m pages Req: Brd:
enough to make all requests good? No! Broadcast may be before the request no bad requests only if two requests at the same time have different colors discrepancy of 3-permutation system is 1 How to Make All Requests Good in an Integral Schedule? P1P1 P1P1 P2P2 P2P2 P3P3 P3P3 P4P4 P4P4 P5P5 P5P5 P6P6 P6P6 P7P7 P7P Req: Brd:
suppose disc χ (π i ) = d π i =(1, 10, 2, 6, 8, 7, 3, 11, 5, 12, 4, 9) χ =(1, 10, 2, 6, 8, 7, 3, 11, 5, 12, 4, 9) order of red elements (1,6,3,5,4,9) right rotate by d-1=1 positions: (9,1,6,3,5,4) broadcast according to this ordering in P 2i-1 #bad quests = d-1 Small Discrepancy Few Bad Requests requests = broadcasts = broadcast after request : good broadcast before request : bad d = 2
“discrepancy” = average discrepancy of l permutations size of BS instance is exponential in l lengths of forbidden intervals grow exponentially Remarks P1P1 P1P1 P2P2 P2P2 P3P3 P3P3 P4P4 P4P4 P5P5 P5P5 P6P6 P6P6 P7P7 P7P7 request good bad
Introduction Discrepancy Problem Broadcast Scheduling Problem Our Results and Techniques Negative Results O(log 1.5 n)-Approximation Outline
A R m×n, x [0,1] n, b=Ax, λ 1, λ 2, …, λ m s.t. output: y [0,1] n, s.t. ½ fraction of coordinates in y are integral Lovett-Meka Framework A A x x b b ×= m n A A y y b b ×= m n ±λ 1 ||A 1 || ±λ 2 ||A 2 || ±λ 3 ||A 3 ||... ±λ m ||A m || “error”
we may broadcast more than 1 page at a time slot tentative schedule of backlog b valid schedule, with additive b loss in the average response time backlog discrepancy Tentative Scheduling 6 time slots, 11 broadcast, backlog = 5
assumptions: fractional schedule is ½-intergal every page is broadcast ≤ Δ = O(log n) times # timeslots ≤ 2Δ × n locally consistent distributions Goal with probability 1/2
Locally Consistent Distribution t f(t) = # broadcasts of p by time t s 1+s 2+s 3+s broadcast p at time 0, 1, 4, 5 randomly select a s (0,1) broadcast at time f -1 (s), f -1 (1+s), f -1 (2+s),…… call (0,1,4,5) a shift for page p
Interesting Intervals # time slots ≤ 2Δ × n “error” repeat log n times : backlog = O(log 3/2 n) 64Δ …… λ = 0 λ = 1 λ = 2 …
previous bestour results approximationO(log 2 n)O(log 3/2 n) integrality gap1 + tiny constΩ(log n) hardnessNP-hardΩ(log 1/2 n) Summary Open problems hardness for 3-permutation(implying the same hardness for broadcast scheduling) discrepancy of l-permutation?