Online Judge上陪审团选人问题用Java实现的一个AC解
原问题位于：http://poj.org/problem?id=1015
以下为问题描述的摘录：
In Frobnia, a far-away country, the verdicts in court trials are determined by a jury consisting of members of the general public. Every time a trial is set to begin, a jury has to be selected, which is done as follows. First, several people are drawn randomly from the public. For each person in this pool, defence and prosecution assign a grade from 0 to 20 indicating their preference for this person. 0 means total dislike, 20 on the other hand means that this person is considered ideally suited for the jury.
Based on the grades of the two parties, the judge selects the jury. In order to ensure a fair trial, the tendencies of the jury to favour either defence or prosecution should be as balanced as possible. The jury therefore has to be chosen in a way that is satisfactory to both parties.
We will now make this more precise: given a pool of n potential jurors and two values di (the defence's value) and pi (the prosecution's value) for each potential juror i, you are to select a jury of m persons. If J is a subset of {1,..., n} with m elements, then D(J ) = sum(dk) k belong to J
and P(J) = sum(pk) k belong to J are the total values of this jury for defence and prosecution.
For an optimal jury J , the value |D(J) - P(J)| must be minimal. If there are several jurys with minimal |D(J) - P(J)|, one which maximizes D(J) + P(J) should be selected since the jury should be as ideal as possible for both parties.
You are to write a program that implements this jury selection process and chooses an optimal jury given a set of candidates.
分析： 每一个candidate可以有被选和未被选两种可能。设O(p,q,diff,sum)表示: 在状态(diff,sum)下，从1, 2, ... p个候选人中选择q个作为陪审团成员的一个最优解, 如果
1）p被选中，那么剩下的q-1个候选人从p-1中选取的方案肯定也是最优的，即O(p,q,diff,sum) = O(p-1,q-1, diff + Ppscore - Pdscore, sum + Ppscore + Pdscore);
2）同样，如果p未被选中，那么剩下的q个候选人从p-1中选取的方案肯定也是最优的，即O(p,q,diff,sum) = O(p-1,q,diff,sum)。
这样最优解就是从这两种情况中选一个符合条件的最佳方案。
这里O(k,0) = {}; O(k,k)={1，2，...，k}
因此符合动态规划(DP)的最优原则。
在计算O(i,j,diff,sum)时并不需要考虑从n, n-1, ..., i+1的组合辩控和（即变量sum)，因为如果有sum1>sum2，那么X + sum1 > X + sum2，而这里X就是从n, n-1, ..., i+1的组合辩控和。当然diff状态必须考虑，因为最终的最优解要求是绝对值，而不是简单和。本人一开始考虑sum，走了一些弯路。
计算顺序举例：如果在动态规划过程(遵循从1,2,...i计算顺序)计算到(i,j)-即从1...i个候选人中选择j个人作为陪审员时，当前状态 i=10,j=6,diff=-5（即代码中的s值: 可变范围diffMin~diffMax），pi=7（控方给第i个候选人的打分）,di=4（辩方给第i个候选人的打分），此时有两个可能最优方案 i=9,j=6,diff=-5（即不选i，记为O(9，6，-5)=代码中的变量prevDs2）和i=9,j=5,diff=-5+(pi- di)=-2（即选i，记为O(9，5，-2)=代码中的变量prevDs1），接下来就要根据这两个可能解的|Diff(1, 2, ...,i)|最小以及Sum(1, 2, ...,i)最大决定取那个解作为最优方案，如果O(9，6，-5)和O(9，5，-2)的最优diff都等于-3，则需要比较下面两个辩控和： 1）第i （i=10）个候选人的辩控和dpiSum + O(9，5，-2)从1,2...9的辩控和prevDs1.sum； 2）O(9，6，-5)从1,2...9的辩控和prevDs1.sum。这里第二个方案中没有包含i。
import java.util.ArrayList;
import java.util.Collections;
import java.util.Comparator;
import java.util.LinkedList;
import java.util.List;
import java.util.Scanner;
/*** Author: ljs* 2011 - 05 - 14* * see also: */
public class Main {class DS{//the global optimal diff of both parties (prosecution and defense)//for the state (i,j,diff)short d; //the current sum from 1 to ishort sum;//the candidates(1~200): format: (1)(2)....(200)byte[] candNrs; }class OptimalChoice { DS[] diffTable;//the offset of the diff tableint offsetDiff; public OptimalChoice(DS[] diffTable,int offsetDiff){this.diffTable = diffTable;this.offsetDiff = offsetDiff;}}class Pair{short nr;short val;Pair(short nr,short val){this.nr =nr;this.val = val;}} private OptimalChoice[] currLevel;private short[] P,D;//private short n;private short m;private int round;public void selectJury(short[] Ps, short[] Ds, short n, short m) {if (n < m || n != Ps.length || Ps.length != Ds.length)return;this.P = new short[n + 1];System.arraycopy(Ps, 0, P, 1, n);this.D = new short[n + 1];System.arraycopy(Ds, 0, D, 1, n); //this.n = n;this.m = m;List<Pair> diffPairs = new LinkedList<Pair>();for(int i=1;i<=n;i++){diffPairs.add(new Pair((short)i, (short)(P[i] - D[i])));}Collections.sort(diffPairs,new Comparator<Pair>(){public int compare(Pair o1,Pair o2){if(o1.val>o2.val)return 1;else if(o1.val==o2.val)return 0;elsereturn -1;}});//Runtime rt = Runtime.getRuntime();OptimalChoice[] lastLevel = null;for (short i = 1; i <= n; i++) {currLevel = new OptimalChoice[m+1];int remainCandidatesCount = n-i; int removeIdx = -1;for(int k=0;k<diffPairs.size();k++){Pair p = diffPairs.get(k);if(p.nr == i){removeIdx = k;break;}} diffPairs.remove(removeIdx); for (short j = 1; j <= m; j++) {int remainJurorsCount = m-j;if (i >= j && (remainCandidatesCount>=remainJurorsCount)) { short diffMin = 0;short diffMax = 0;int u=0,v=diffPairs.size()-1;int cnt = 0;while(cnt<remainJurorsCount){diffMin += diffPairs.get(u).val;diffMax += diffPairs.get(v).val;u++; v--;cnt++;} short OFFSET_DIFF = (short)-diffMin; short diffDim = (short)(diffMax - diffMin + 1); DS[] diffTable = new DS[diffDim]; currLevel[j]=new OptimalChoice(diffTable,OFFSET_DIFF);//only one choice: select all candidates from 1...i if (i == j) { if(n-i==m-j && i<n){//go on to upper levelbreak; }else{short diffTotal = 0;short sumTotal = 0;for(int k=1;k<=i;k++){diffTotal += P[k] - D[k];sumTotal += P[k] + D[k];} for(short s=diffMin;s<=diffMax;s++){//update the current table DS dsNow = new DS();dsNow.d = (short)(diffTotal + s);dsNow.sum = (short)sumTotal;//dsNow.selected = true;//dsNow.selectAllPrev = true;byte[] candNrs = new byte[i];for(byte k=1;k<=i;k++){candNrs[k-1] = k;} dsNow.candNrs = candNrs;//update the current tablediffTable[s + OFFSET_DIFF] = dsNow; }} } else {/*O(i,j,diff,sum) = minabsOrmaxtotal(O(i-1,j-1,diff+pi-di), O(i-1,j,diff));if diff is equal, check the max in (sum_i + O(i-1,j-1).sum, O(i-1,j).sum)*/ // case 1: select iOptimalChoice prevOptimalChoice1 = lastLevel[j - 1];int dpiDiff1 = P[i] - D[i];int dpiSum = P[i] + D[i];//case 2: not select iOptimalChoice prevOptimalChoice2 = lastLevel[j];for(short s=diffMin;s<=diffMax;s++){DS[] diffTable2 = prevOptimalChoice2.diffTable; //get the prev optimal valueDS prevDs2 = diffTable2[s + prevOptimalChoice2.offsetDiff];int diffNew1 = dpiDiff1 + s; DS[] diffTable1 = null;if(prevOptimalChoice1 != null)diffTable1 = prevOptimalChoice1.diffTable;DS prevDs1 = null;if(diffTable1 != null)prevDs1 = diffTable1[diffNew1 + prevOptimalChoice1.offsetDiff];int diff1 = 0; int sum1 = 0;if(prevDs1 ==null){diff1 = Math.abs(diffNew1);sum1 = dpiSum;}else{diff1 = Math.abs(prevDs1.d); sum1 = dpiSum + prevDs1.sum;}//this case can't be possible//if(prevDs2 ==null){ // diff2 = Math.abs(s);// sum2 = 0;//}else{int diff2 = Math.abs(prevDs2.d); int sum2 = prevDs2.sum;//}int diffNow = s + OFFSET_DIFF;DS dsNow = new DS();boolean choice1 = true;if (diff1 < diff2) {choice1 = true; }else if (diff1 > diff2) {// select choice #2choice1 = false; }else { if(sum1 >= sum2){choice1 = true;}else{// select choice #2choice1 = false;} }if(choice1){// select choice #1 if(prevDs1 == null){dsNow.d = (short)diffNew1;}else{dsNow.d = prevDs1.d; }dsNow.sum = (short)sum1;//dsNow.selected = true; byte[] candNrs = null;if(prevDs1 != null){candNrs = new byte[prevDs1.candNrs.length+1]; System.arraycopy(prevDs1.candNrs, 0, candNrs, 0, prevDs1.candNrs.length); candNrs[candNrs.length-1] = (byte)i;}else{candNrs = new byte[1];candNrs[0]= (byte)i;}dsNow.candNrs = candNrs; }else{//if(prevDs2 == null){// dsNow.d = s;//}else{dsNow.d = prevDs2.d; //}dsNow.sum = (short)sum2; if(prevDs2 != null){byte[] candNrs = new byte[prevDs2.candNrs.length]; System.arraycopy(prevDs2.candNrs, 0, candNrs, 0, prevDs2.candNrs.length); dsNow.candNrs = candNrs;} }diffTable[diffNow] = dsNow; } } } // i < j is invalid (e.g. n<m is not valid)//remove the last level's items from 1 to j-2: if(lastLevel!= null && j>=1 && lastLevel[j-1] != null) { DS[] diffTable = lastLevel[j-1].diffTable;for(int row=0;row<diffTable.length;row++){ diffTable[row] = null; }//remove the DS items lastLevel[j-1].diffTable = null;lastLevel[j-1] = null; }// Collect garbage //rt.gc();}//long availMem = rt.freeMemory()/1000000;//System.out.format("%d mb, i=%d%n", availMem,i);lastLevel = null;lastLevel = currLevel;//rt.gc();//System.gc();} }public void setRound(int round) {this.round = round;}public void output(){OptimalChoice juryChoiceTable = currLevel[m];DS[] diffSumTable = juryChoiceTable.diffTable;int diff = 0;int offsetDiff = 0; DS optimal = diffSumTable[diff + offsetDiff];int totalP=0;int totalD=0;byte[] candNrs = optimal.candNrs;List<Short> jurors = new ArrayList<Short>();for(int i=0;i<candNrs.length;i++){short candidateNr = (short)(candNrs[i] & 0x00ff);jurors.add(candidateNr);totalP += P[candidateNr];totalD += D[candidateNr];};Collections.sort(jurors);System.out.println("Jury #"+this.round);System.out.println("Best jury has value "+totalP + " for prosecution and value " + totalD +" for defence:");for(short jurorNr:jurors){System.out.print(" "+jurorNr);} System.out.println();System.out.println();}public static void main(String[] args) throws Exception {List<short[]> PS = new ArrayList<short[]>();List<short[]> DS = new ArrayList<short[]>();List<Short> NS = new ArrayList<Short>();List<Short> MS = new ArrayList<Short>();short[] P=null;short[] D=null;short n=0,m=0;boolean jobsAvail = false;try{Scanner cin=new Scanner(System.in);String line = cin.nextLine(); if(line != null) { boolean nextRound=true;int candidateLine=0;//int round=1;do{ line = line.trim(); if(line.length()==0){if(P != null && n >0){ PS.add(P);DS.add(D);NS.add(n);MS.add(m);jobsAvail = true;//round++;}//do the next roundnextRound = true; P = null;D = null;}else{ short[] pair = new short[]{-1,-1};String[] tokens = line.split(" ");for(int i=0,j=0;i<tokens.length;i++){String token = tokens[i].trim();if(token.length()>0 && j<2){pair[j++] = Short.parseShort(token);}}//endif(pair[0]==0 && pair[1]==0){ if(candidateLine<n){P[candidateLine] = pair[0];D[candidateLine++]=pair[1];}else{if(P != null){PS.add(P);DS.add(D);NS.add(n);MS.add(m);jobsAvail = true;}break; }}else{if(nextRound){nextRound = false;n = pair[0];m = pair[1];P = new short[n];D = new short[n];candidateLine=0;}else{if(candidateLine<n){P[candidateLine] = pair[0];D[candidateLine++]=pair[1];}}} } line = cin.nextLine();}while(line !=null);}}catch(Exception e){throw e;}if(jobsAvail){ for(int i=0;i<PS.size();i++){Main js = new Main();js.setRound(i+1);js.selectJury(PS.get(i), DS.get(i), NS.get(i),MS.get(i));js.output();}}}
}
测试：
4 2 // sample input
1 2
2 3
4 1
6 2
5 3 // prosecution > defense
13 11
3 17
15 20
6 13
17 9
8 5 // prosecution < defense
3 5
17 16
6 0
17 10
6 14
3 19
4 13
0 17
8 4 // prosecution == defense
19 18
5 9
0 9
5 0
4 4
16 9
4 15
20 5
2 1 // two solutions with diffence 1, but one of them
1 2 // has a larger value (version 1)
4 3
2 1 // two solutions with diffence 1, but one of them
4 3 // has a larger value (version 2)
1 2
2 1 // two solutions with diffence 1, but one of them
3 4 // has a larger value (version 3)
1 2
2 1 // two solutions with diffence 1, but one of them
2 1 // has a larger value (version 4)
3 4
18 7 // only differences +1,0,-1 occur with diffenent
1 1 // sums (of values). Make sure that we are really
1 2 // selecting the maximum sum.
2 1
10 10
11 10
10 11
7 7
7 8
8 7
14 14
14 15
15 14
4 4
4 5
5 4
19 19
19 20
20 19
21 13 // all possible pairs of equal values: (0,0),...(20,20)
20 20 // again, there are many solutions with difference 0, but
19 19 // we want the one with maximum total value!
18 18
14 14
13 13
5 5
4 4
3 3
12 12
10 10
9 9
8 8
17 17
16 16
15 15
7 7
6 6
11 11
2 2
1 1
0 0
10 10 // n = m
8 16
9 6
6 6
2 15
0 13
17 8
17 3
13 13
15 10
18 2
10 9 // n = m+1
12 13
16 17
2 7
1 9
5 7
10 9
9 8
9 19
2 10
3 19
10 8 // n = m+2
14 4
18 17
20 12
10 16
8 14
1 16
9 10
16 11
4 2
10 16
10 5 // partition problem: note that the values of
3 8 // both parties add up to 80. There exits a solution
15 8 // with diffence 0 iff the prosections values can be
11 8 // divided in two subsets with sum 40 each. This
7 8 // example shows how to reduce BI-PARTITION to this
1 8 // problem and therefore show its NP-completeness.
17 8
8 8
2 8
13 8
3 8
1 1 // now for the boundary test: first comes n=m=1
9 11
1 1
7 3
1 1
5 5
1 1
0 0
1 1
0 20
1 1
20 0
1 1
20 20
20 20 // now comes n=m=20.
0 0 // first 20 times (0,0), then (0,20), (20,0) and (20,20)
0 0 // finally a random sample
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
20 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
20 20
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
7 20
10 1
14 17
11 13
3 5
19 19
1 16
4 18
9 1
13 3
7 16
6 3
1 20
4 5
6 13
6 8
10 9
11 11
0 7
12 15
200 20 // and now for maximum values: n = 200, m = 20.
0 0 // again a five test series as above
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
200 20
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
20 0
200 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
0 20
200 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
20 20
200 20
12 20
10 0
13 17
15 1
8 12
4 12
9 5
3 3
13 18
13 6
12 11
10 6
3 1
6 1
20 2
20 18
19 2
12 12
1 8
12 6
6 11
11 16
11 0
2 1
7 6
15 6
13 7
16 17
5 11
0 14
17 8
10 9
12 14
9 9
20 2
13 4
5 19
17 0
2 12
13 8
4 17
5 3
11 19
6 9
1 3
5 10
18 0
6 6
2 8
20 11
0 8
5 1
12 17
17 8
9 12
9 10
3 0
0 19
20 9
15 16
4 6
1 1
19 13
9 17
0 8
1 4
19 14
1 6
16 14
12 12
19 18
19 19
3 17
9 17
4 6
16 19
14 3
9 4
13 19
3 16
19 9
0 2
8 9
10 10
15 3
16 2
16 4
12 6
13 0
8 0
11 8
18 15
18 18
2 20
7 2
14 1
3 0
0 15
11 16
0 18
1 17
11 11
9 9
19 6
6 1
18 13
19 8
9 0
2 18
16 9
18 12
10 13
12 13
16 5
7 17
5 4
13 8
0 4
8 19
12 14
0 2
6 19
9 0
18 6
7 19
18 19
15 7
9 12
12 5
0 1
15 15
1 16
20 16
10 10
11 2
7 4
16 12
16 0
4 3
17 18
8 14
13 7
5 19
14 20
19 15
5 19
18 7
6 9
0 2
7 3
11 14
7 0
8 5
18 15
0 2
11 2
11 17
11 8
5 16
5 13
12 12
8 6
20 14
13 14
0 1
10 15
12 17
14 15
2 6
17 14
6 13
10 10
13 6
6 16
15 11
4 5
15 18
11 13
19 20
2 0
12 9
14 2
10 20
0 7
16 2
16 13
20 17
14 6
5 15
7 19
17 10
0 19
13 15
17 18
19 9
14 1
15 2
8 8
2 15
20 2
200 10 // and another large test, n=200,m=10
16 11
9 10
2 19
2 18
8 7
18 3
9 16
10 3
16 3
9 16
4 14
10 0
10 13
7 1
5 3
4 19
9 1
7 4
19 20
10 5
19 13
9 5
12 4
18 12
4 12
0 8
3 0
10 18
6 19
3 3
17 8
12 17
5 3
16 4
4 10
8 18
17 17
18 9
6 0
20 6
8 13
8 2
16 17
18 7
14 0
0 3
14 0
10 11
9 12
7 9
1 8
16 19
20 18
14 19
17 11
16 3
11 11
5 1
19 4
0 20
12 18
19 18
7 16
9 2
17 17
5 7
6 12
9 18
17 4
6 7
20 7
10 6
14 3
13 13
16 15
6 0
19 7
15 6
16 19
20 17
17 0
12 18
7 6
16 8
5 17
10 15
1 5
4 19
1 6
12 6
4 10
5 9
16 5
8 17
4 3
1 6
6 5
3 19
13 6
11 1
4 9
1 19
6 3
10 11
13 8
0 13
5 9
1 10
13 19
15 0
1 10
13 9
6 13
16 3
2 17
13 20
12 7
10 18
12 12
15 18
4 18
14 10
17 18
11 2
5 2
15 15
19 14
12 5
18 19
19 12
5 17
9 20
12 0
19 13
14 7
2 16
7 11
15 15
8 6
8 2
1 13
18 15
1 2
17 20
4 19
9 16
10 12
7 12
9 3
5 14
0 12
14 10
1 11
12 18
16 16
10 15
7 8
9 10
13 7
11 14
6 3
7 20
16 4
17 1
6 7
12 11
18 1
6 19
10 15
15 12
1 17
12 3
13 16
14 14
15 1
18 8
19 15
6 15
12 9
14 16
7 7
11 0
1 20
16 13
15 3
5 13
9 11
12 15
20 16
2 17
1 19
17 15
12 5
16 2
2 2
19 9
15 0
20 11
18 12
14 5
0 0
Jury #1
Best jury has value 6 for prosecution and value 4 for defence:
2 3
Jury #2
Best jury has value 36 for prosecution and value 33 for defence:
1 4 5
Jury #3
Best jury has value 50 for prosecution and value 53 for defence:
2 3 4 5 7
Jury #4
Best jury has value 33 for prosecution and value 33 for defence:
2 5 7 8
Jury #5
Best jury has value 4 for prosecution and value 3 for defence:
2
Jury #6
Best jury has value 4 for prosecution and value 3 for defence:
1
Jury #7
Best jury has value 3 for prosecution and value 4 for defence:
1
Jury #8
Best jury has value 3 for prosecution and value 4 for defence:
2
Jury #9
Best jury has value 111 for prosecution and value 111 for defence:
4 10 11 12 16 17 18
Jury #10
Best jury has value 182 for prosecution and value 182 for defence:
1 2 3 4 5 9 10 11 12 13 14 15 18
Jury #11
Best jury has value 105 for prosecution and value 92 for defence:
1 2 3 4 5 6 7 8 9 10
Jury #12
Best jury has value 66 for prosecution and value 99 for defence:
1 2 3 4 5 6 7 8 9
Jury #13
Best jury has value 93 for prosecution and value 94 for defence:
1 2 3 4 6 8 9 10
Jury #14
Best jury has value 40 for prosecution and value 40 for defence:
2 4 8 9 10
Jury #15
Best jury has value 9 for prosecution and value 11 for defence:
1
Jury #16
Best jury has value 7 for prosecution and value 3 for defence:
1
Jury #17
Best jury has value 5 for prosecution and value 5 for defence:
1
Jury #18
Best jury has value 0 for prosecution and value 0 for defence:
1
Jury #19
Best jury has value 0 for prosecution and value 20 for defence:
1
Jury #20
Best jury has value 20 for prosecution and value 0 for defence:
1
Jury #21
Best jury has value 20 for prosecution and value 20 for defence:
1
Jury #22
Best jury has value 0 for prosecution and value 0 for defence:
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Jury #23
Best jury has value 0 for prosecution and value 400 for defence:
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Jury #24
Best jury has value 400 for prosecution and value 0 for defence:
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Jury #25
Best jury has value 400 for prosecution and value 400 for defence:
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Jury #26
Best jury has value 154 for prosecution and value 220 for defence:
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Jury #27
Best jury has value 0 for prosecution and value 0 for defence:
181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200
Jury #28
Best jury has value 400 for prosecution and value 0 for defence:
181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200
Jury #29
Best jury has value 0 for prosecution and value 400 for defence:
181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200
Jury #30
Best jury has value 400 for prosecution and value 400 for defence:
181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200
Jury #31
Best jury has value 350 for prosecution and value 350 for defence:
1 16 28 60 71 72 76 92 93 126 133 140 144 145 154 163 177 179 187 194
Jury #32
Best jury has value 182 for prosecution and value 182 for defence:
19 43 53 62 79 80 123 129 144 189
转载于:https://www.cnblogs.com/ljsspace/archive/2011/06/05/2073355.html
Online Judge上陪审团选人问题用Java实现的一个AC解相关推荐
- 这可能是史上功能最全的Java权限认证框架！
点击关注公众号,Java干货及时送达 今天给大家推荐的这个开源项目超级棒,可能是史上功能最全的 Java 权限认证框架! 这个开源项目就是:sa-token . Sa-Token是什么? sa-tok ...
- 历史上最简单的一道Java面试题，但无人能通过
作者:方志宏 来源:zhuanlan.zhihu.com/p/57859872 这可能是历史上最简单的一道java面试题了. 题目很简单,完成代码,判断一个整数是否是奇数: public boolea ...
- 史上最难10道Java面试题！
点击上方"朱小厮的博客",选择"设为星标" 回复"1024"获取独家整理的学习资料 这是我收集的10个最棘手的Java面试问题列表.这些问题 ...
- java代码上传exel,excle上传服务器并解析!求excel上传到服务器的java代码
如何将数据上传给服务器 医嘱以形式发送过来? 办法有很多,最简单的,就是在机器里,建立2个数据库A,B,假如A是外务器. 在数据库中,以A数据库做发布,让B数据库订阅,弄好以后,A数据库的数据就会自动 ...
- 史上最全的中高级JAVA工程师-面试题汇总
了解博主本人,请阅读<成就一亿技术人,我在CSDN的这九年> 文档下载地址: https://download.csdn.net/download/shengqianfeng/120804 ...
- springboot毕设项目线上跳蚤市场平台iy7e7（java+VUE+Mybatis+Maven+Mysql）
springboot毕设项目线上跳蚤市场平台iy7e7(java+VUE+Mybatis+Maven+Mysql) 项目运行 环境配置: Jdk1.8 + Tomcat8.5 + Mysql + HB ...
- 在Mac上使用idea搭建flink java开发环境
1.环境 本文档记录的是使用flink的java API简单地创建应用的过程. 前置条件:需要安装Java.maven和flink. 1.1 Java环境变量 Java需要jdk.path.class ...
- 先搞清楚这些问题，简历上再写你熟悉Java！
原创声明 本文作者:黄小斜 转载请务必在文章开头注明出处和作者. 系列文章介绍 本文是<五分钟学Java>系列文章的一篇 本系列文章主要围绕Java程序员必须掌握的核心技能,结合我个人三年 ...
- GitHub上最火的两份Java面试小册，Star已经超百万
大家好,给大家带来两份GitHub上Star超百万的Java面试小册,独家整理!! Java面试小册内容涵盖Java.MyBatis.ZooKeeper.Dubbo.Elasticsearch.Mem ...
最新文章
- VS2015 Qt5
- postgresql 数据库 客户端认证
- GDI 绘制圆角矩形
- python3 copy_python3 深浅copy对比详解
- Logistic回归 python实现
- ES6 let 和 const 关键字
- 关于kafka的几个问题
- 关于mysql中select出数据未排序问题
- http请求后status为canceled处理
- linux 0.11根文件系统,构建一个最小Linux根文件系统
- 计算机技术中采用二进制,在计算机技术中采用二进制.docx
- c++ 11/14新特性
- 【蓝桥杯单片机笔记6】PWM脉宽调制信号
- 实现远程连接阿里云的数据库
- UITextView 首行缩进
- Excel自动生成条码工具
- C++使电脑强制关机
- ansys 服务器无响应,ANSYS出现的各种错误及解决方法
- 不会就要问，求大神解决一下安装linux不认内置硬盘的问题
- 【JQ _DOM】DOM
热门文章
- fiddler使用技巧进阶，如何抓包修改数据？——AutoResponder重定向
- php自动计算增长率,如何写sql计算增长率？
- c语言中void跟argv,argc和argv []在C语言中
- 量子计算机 漫画,漫画 | 10分钟看懂量子比特、量子计算和量子算法
- Python培训教程：什么是Python全局解释器锁(GIL)?
- html5地理定位数据
- android TextView里边实现图文混配效果
- javascript通过json数据按格式生成一个按字母分类排序的分类信息表
- 【android】java.lang.NoClassDefFoundError或classnotfount等异常错误
- 《一江春水向东流》之随笔