3098. Find the Sum of Subsequence Powers
Read the full problem statement on LeetCode.
Difficulty: hard Acceptance: 23% Topics: Array, Dynamic Programming, Sorting
View full problem on LeetCode Reading material
Reference solution (spoiler · python)
# Time: O(n^2 + len(diffs) * n * k) = O(n^3 * k) at most
# Space: O(len(diffs) + n * k) = O(n^2) at most
# sort, dp, prefix sum, two pointers
class Solution(object):
def sumOfPowers(self, nums, k):
MOD = 10**9+7
nums.sort()
result = prev = 0
for mn in sorted({nums[j]-nums[i] for i in xrange(len(nums)) for j in xrange(i+1, len(nums))}, reverse=True):
dp = [[0]*(k+1) for _ in xrange(len(nums)+1)]
dp[0][0] = 1
j = 0
for i in xrange(len(nums)):
j = next((j for j in xrange(j, len(nums)) if nums[i]-nums[j] < mn), len(nums))
for l in xrange(1, k+1):
dp[i+1][l] = (dp[i+1][l]+dp[(j-1)+1][l-1])%MOD # dp[i+1][l]: count of subsequences of length l ending at i having min diff >= mn
for l in xrange(k+1):
dp[i+1][l] = (dp[i+1][l]+dp[i][l])%MOD # dp[i+1][l]: accumulated count of subsequences of length l ending at [0, i] having min diff >= mn
cnt = (dp[-1][k]-prev)%MOD
result = (result+mn*cnt)%MOD
prev = dp[-1][k]
return result
# Time: O(n^3 * len(diffs)) = O(n^5) at most
# Space: O(n^2 * len(diffs)) = O(n^4) at most
import collections
# sort, dp
class Solution2(object):
def sumOfPowers(self, nums, k):
"""
:type nums: List[int]
:type k: int
:rtype: int
"""
MOD = 10**9+7
nums.sort()
dp = [[collections.defaultdict(int) for _ in xrange(len(nums)+1)] for _ in xrange(len(nums))]
for i in xrange(len(nums)):
for j in xrange(max(k-(len(nums)-i+1)-1, 0), i):
diff = nums[i]-nums[j]
dp[i][2][diff] += 1
for l in xrange(max(k-(len(nums)-i+1), 0), i+1):
for mn, cnt in dp[j][l].iteritems():
dp[i][l+1][min(diff, mn)] = (dp[i][l+1][min(diff, mn)]+cnt)%MOD
return reduce(lambda accu, x: (accu+x)%MOD, ((mn*cnt)%MOD for i in xrange(k-1, len(dp)) for mn, cnt in dp[i][k].iteritems()))
Solution from kamyu104/LeetCode-Solutions · MIT