Tutorial: The fastest tree algorithm, the XOR Linked Tree

Revision en30, by pajenegod, 2024-10-18 23:05:10

Hi Codeforces!

This is a blog about a really cool folklore technique that seems to have been forgotten over time. I currently only know of one place where it is mentioned on the internet, https://www.algonotes.com/en/minimal-memory-dp/, which is a tutorial for a problem called Matchings https://dmoj.ca/problem/ontak2010sko from a POI training camp in 2010. But I've heard that this technique is even older than that. I originally learnt of this technique a couple of years back when I was digging through the fastest submissions on 321-C - Ciel the Commander and I saw it being used in a submission 9076307 from 2014. I don't know if there is a name for this technique, so I've decided to call it the XOR Linked Tree because of its close connection to XOR linked lists https://en.wikipedia.org/wiki/XOR_linked_list.

The XOR Linked Tree is a technique for solving tree problems using very little time and memory. For example, every fastest solution in the "Tree Algorithms" section on https://cses.fi/problemset/ currently uses this technique. So it is record breaking fast!

The idea:

Suppose in the input of a problem, you are given a tree as $$$n - 1$$$ edges. To construct the XOR Linked Tree data-structure from this, you go through all of the edges and compute the following two things for every node in the tree:

  1. deg[node] = The degree of the node.
  2. xor[node] = The XOR of all neighbours of the node.

Storing this will only take $$$2 n$$$ integers, so this is very cheap memory wise compared to using an adjacency list. It is also really fast, since there are no expensive .push_back/.append calls.

Claim: Using deg and xor, it is possible to reconstruct the tree.

Proof: Identify a leaf node in the tree (this can be done by finding a node with degree 1 in deg). Note that a leaf only has one neighbour, so the XOR of all of its neighbours is simply that neighbour. I.e. the neighbour of a leaf is xor[leaf]. Now remove this leaf from the tree ("pluck it") by lowering the degree of its neighbour by one (deg[neighbour] -= 1) and XORing away the leaf from xor[neighbour] (xor[neighbour] ^= leaf). Now we can just repeat this process of plucking leaves until there a single node left. We have now reconstructed the original tree.

Here is an implementation of the process described in the proof above ^

for i in range(n):
    node = i
    # While node is a leaf
    while deg[node] == 1:
        nei = xor[node]

        # Pluck away node
        deg[node] = 0
        deg[nei] -= 1
        xor[nei] ^= node

        # Check if nei has become a leaf
        node = nei

Something to note is that it is possible to control which node is left at the end (the "root") by setting its degree to 0 before running the algorithm. So the final XOR Linked Tree traversal algorithm is

deg[root] = 0
for i in range(n):
    node = i
    # While node is a leaf
    while deg[node] == 1:
        nei = xor[node]

        # Pluck away node
        deg[node] = 0
        deg[nei] -= 1
        xor[nei] ^= node

        # Check if nei has become a leaf
        node = nei

This is it! This is the entire algorithm! The deg list is now completely 0 and the xor list now contains the parent of every node.

Discussion

This tree traversal algorithm is different compared to BFS or DFS. For one, it doesn't use anything like a queue or a stack. Instead it simply removes one leaf at a time using a while loop inside of a for loop. The order you traverse the nodes in is kind of like doing a BFS in reverse, but it is not exactly the same thing.

Discussion (Advanced)

One last thing I want to mention is that on some tree problems, you specifically need a DFS ordering. For example, there are tree problems on cses.fi that requires you to compute LCA. Most (if not all) algorithms/data-structures for computing LCA are based on DFS. So is it still possible to use XOR Linked Tree traversal algorithm for these kind of problems?

The answer is yes, it is possible to use the XOR Linked Tree traversal algorithm to compute a DFS ordering. The trick is to first compute the sizes of all subtrees, and then use this information to compute the index of each node in a DFS ordering.

# Store the order of the plucked nodes in the XOR Linked Tree traversal
XOR_traversal_order = []

deg[root] = 0
for i in range(n):
    node = i
    # While node is a leaf
    while deg[node] == 1:
        nei = xor[node]

        # Pluck away node
        XOR_traversal_order.append(node)
        deg[node] = 0
        deg[nei] -= 1
        xor[nei] ^= node

        # Check if nei has become a leaf
        node = nei

# Compute subtree sizes
subtree_size = [1] * n
for node in XOR_traversal_order:
    # Note that xor[node] is the parent of node
    p = xor[node]
    subtree_size[p] += subtree_size[node]

# Compute the index each node would have in a DFS using subtree_size
# NOTE: This modifies subtree_size
for node in reversed(XOR_traversal_order):
    p = xor[node]
    subtree_size[node], subtree_size[p] = subtree_size[p], subtree_size[p] - subtree_size[node]

DFS_traversal_order = [None] * n
for node in range(n):
    DFS_traversal_order[subtree_size[node] - 1] = node

This might be one of the weirdest ways to do a DFS ever. But it is blazingly fast!

Benchmark

Here are some benchmarks based on https://judge.yosupo.jp/problem/tree_diameter

Python Benchmark Time Memory Submission link
XOR Linked Tree 315 ms 188.93 Mib 243487
BFS with adjency list 1076 ms 227.96 Mib 243489
DFS (stack) with adjency list 997 ms 228.90 Mib 243490
DFS (recursive) with adjency list 2119 ms 731.60 Mib 243496
C++ Benchmark Time Memory Submission link
XOR Linked Tree 31 ms 15.50 Mib 243281
Adjency list (vector) 97 ms 59.54 Mib 243277
Adjency list (basic_string) 89 ms 58.62 Mib 243285
"Chinese adjency list" 73 ms 45.24 Mib 243280
Tags tree, traversal, xor, constant factor, memory limit, dfs, fast

History

 
 
 
 
Revisions
 
 
  Rev. Lang. By When Δ Comment
en39 English pajenegod 2024-10-20 23:13:06 1 Tiny change: 'rently uses this tech' -> 'rently use this tech'
en38 English pajenegod 2024-10-20 23:09:06 9 Tiny change: 's problem called Matchings https://d' -> 's problem "Matchings" https://d'
en37 English pajenegod 2024-10-20 23:08:34 124 Tiny change: '([see this](https://' -> '([see this for more information](https://'
en36 English pajenegod 2024-10-20 18:17:51 10
en35 English pajenegod 2024-10-19 14:16:01 47
en34 English pajenegod 2024-10-19 13:29:01 101 (published)
en33 English pajenegod 2024-10-18 23:22:01 415 Tiny change: 'xicPie9]. A big thanks to [user:' -> 'xicPie9]. Credit to [user:'
en32 English pajenegod 2024-10-18 23:11:09 32 Tiny change: ':9076307] from 201' -> ':9076307] by [user:Marcin_smu] from 201'
en31 English pajenegod 2024-10-18 23:06:04 12
en30 English pajenegod 2024-10-18 23:05:10 550 Tiny change: 'cy list | TBD | TBD | [243496' -> 'cy list | 2119 ms | 731.60 Mib | [243496'
en29 English pajenegod 2024-10-18 22:58:35 639
en28 English pajenegod 2024-10-18 02:53:10 30
en27 English pajenegod 2024-10-18 02:42:28 41
en26 English pajenegod 2024-10-18 02:41:41 4 Tiny change: 'is **yes**. It is possi' -> 'is **yes**, it is possi'
en25 English pajenegod 2024-10-18 02:40:25 4 Tiny change: 'mpared to DFS or BFS. For on' -> 'mpared to BFS or DFS. For on'
en24 English pajenegod 2024-10-18 02:32:59 6 Tiny change: 'mpared to an adjace' -> 'mpared to using an adjace'
en23 English pajenegod 2024-10-18 02:32:30 4 Tiny change: 'hbours of node.`\n\' -> 'hbours of the node.`\n\'
en22 English pajenegod 2024-10-18 02:31:12 7
en21 English pajenegod 2024-10-18 02:18:58 57
en20 English pajenegod 2024-10-18 02:14:41 4 Tiny change: 'Ring away leaf from' -> 'Ring away the leaf from'
en19 English pajenegod 2024-10-18 02:10:09 6 Tiny change: 'ittle time/memory. Fo' -> 'ittle time and memory. Fo'
en18 English pajenegod 2024-10-18 02:09:21 1 Tiny change: 'inked_list .\n\nThe X' -> 'inked_list.\n\nThe X'
en17 English pajenegod 2024-10-18 02:07:45 150
en16 English pajenegod 2024-10-18 01:42:13 10 Tiny change: ' one leaf using a w' -> ' one leaf at a time using a w'
en15 English pajenegod 2024-10-18 01:40:06 4 Tiny change: 'o call it _XOR Link' -> 'o call it the _XOR Link'
en14 English pajenegod 2024-10-18 01:37:10 25
en13 English pajenegod 2024-10-18 01:36:38 20
en12 English pajenegod 2024-10-18 01:35:21 5 Tiny change: 'lems, you need specifica' -> 'lems, you specifica'
en11 English pajenegod 2024-10-18 01:34:37 13 Tiny change: 'different than DFS or BF' -> 'different compared to DFS or BF'
en10 English pajenegod 2024-10-18 01:33:15 84
en9 English pajenegod 2024-10-18 01:31:55 13
en8 English pajenegod 2024-10-18 01:30:52 1 Tiny change: '/`.append`s calls.\n\' -> '/`.append` calls.\n\'
en7 English pajenegod 2024-10-18 01:30:17 13 Tiny change: 'o through the list of edges and' -> 'o through all of the edges and'
en6 English pajenegod 2024-10-18 01:29:29 17 Tiny change: 'st .\n\nThis is a tech' -> 'st .\n\nThe XOR Linked Tree is a tech'
en5 English pajenegod 2024-10-18 01:28:44 16
en4 English pajenegod 2024-10-18 01:26:02 1706 Tiny change: ' I saw it was being use' -> ' I saw it being use'
en3 English pajenegod 2024-10-18 00:49:58 3 Tiny change: ' algorithm. \n\n## Dis' -> ' algorithm!\n\n## Dis'
en2 English pajenegod 2024-10-18 00:46:04 1884 Tiny change: 'e are no `.push_back`/`.append`.\n\n**Cla' -> 'e are no `vector<vector>`s.\n\n**Cla'
en1 English pajenegod 2024-10-18 00:15:45 2085 Initial revision (saved to drafts)