-
On the Group and Color Isomorphism Problems
Authors:
François Le Gall,
David J. Rosenbaum
Abstract:
In this paper, we prove results on the relationship between the complexity of the group and color isomorphism problems. The difficulty of color isomorphism problems is known to be closely linked to the the composition factors of the permutation group involved. Previous works are primarily concerned with applying color isomorphism to bou nded degree graph isomorphism, and have therefore focused on…
▽ More
In this paper, we prove results on the relationship between the complexity of the group and color isomorphism problems. The difficulty of color isomorphism problems is known to be closely linked to the the composition factors of the permutation group involved. Previous works are primarily concerned with applying color isomorphism to bou nded degree graph isomorphism, and have therefore focused on the alternating composit ion factors, since those are the bottleneck in the case of graph isomorphism.
We consider the color isomorphism problem with composition factors restricted to those other than the alternating group, show that group isomorphism reduces in n^(O(log log n)) time to this problem, and, conversely, that a special case of this color isomorphism problem reduces to a slight generalization of group isomorphism. We then sharpen our results by identifying the projective special linear group as the main obstacle to faster algorithms for group isomorphism and prove that the aforementioned reduc tion from group isomorphism to color isomorphism in fact produces only cyclic and projective special linear factors. Our results demonstrate that, just as the alternatin g group was a barrier to faster algorithms for graph isomorphism for three decades, the projective special linear group is an obstacle to faster algorithms for group isomorphism.
△ Less
Submitted 27 September, 2016;
originally announced September 2016.
-
Beating the Generator-Enumeration Bound for Solvable-Group Isomorphism
Authors:
David J. Rosenbaum
Abstract:
We consider the isomorphism problem for groups specified by their multiplication tables. Until recently, the best published bound for the worst-case was achieved by the n^(log_p n + O(1)) generator-enumeration algorithm. In previous work with Fabian Wagner, we showed an n^((1 / 2) log_p n + O(log n / log log n)) time algorithm for testing isomorphism of p-groups by building graphs with degree boun…
▽ More
We consider the isomorphism problem for groups specified by their multiplication tables. Until recently, the best published bound for the worst-case was achieved by the n^(log_p n + O(1)) generator-enumeration algorithm. In previous work with Fabian Wagner, we showed an n^((1 / 2) log_p n + O(log n / log log n)) time algorithm for testing isomorphism of p-groups by building graphs with degree bounded by p + O(1) that represent composition series for the groups and applying Luks' algorithm for testing isomorphism of bounded degree graphs.
In this work, we extend this improvement to the more general class of solvable groups to obtain an n^((1 / 2) log_p n + O(log n / log log n)) time algorithm. In the case of solvable groups, the composition factors can be large which prevents previous methods from outperforming the generator-enumeration algorithm. Using Hall's theory of Sylow bases, we define a new object that generalizes the notion of a composition series with small factors but exists even when the composition factors are large. By constructing graphs that represent these objects and running Luks' algorithm, we obtain our algorithm for solvable-group isomorphism. We also extend our algorithm to compute canonical forms of solvable groups while retaining the same complexity.
△ Less
Submitted 1 December, 2014;
originally announced December 2014.
-
Beating the Generator-Enumeration Bound for $p$-Group Isomorphism
Authors:
David J. Rosenbaum,
Fabian Wagner
Abstract:
We consider the group isomorphism problem: given two finite groups G and H specified by their multiplication tables, decide if G cong H. For several decades, the n^(log_p n + O(1)) generator-enumeration bound (where p is the smallest prime dividing the order of the group) has been the best worst-case result for general groups. In this work, we show the first improvement over the generator-enumerat…
▽ More
We consider the group isomorphism problem: given two finite groups G and H specified by their multiplication tables, decide if G cong H. For several decades, the n^(log_p n + O(1)) generator-enumeration bound (where p is the smallest prime dividing the order of the group) has been the best worst-case result for general groups. In this work, we show the first improvement over the generator-enumeration bound for p-groups, which are believed to be the hard case of the group isomorphism problem. We start by giving a Turing reduction from group isomorphism to n^((1 / 2) log_p n + O(1)) instances of p-group composition-series isomorphism. By showing a Karp reduction from p-group composition-series isomorphism to testing isomorphism of graphs of degree at most p + O(1) and applying algorithms for testing isomorphism of graphs of bounded degree, we obtain an n^(O(p)) time algorithm for p-group composition-series isomorphism. Combining these two results yields an algorithm for p-group isomorphism that takes at most n^((1 / 2) log_p n + O(p)) time. This algorithm is faster than generator-enumeration when p is small and slower when p is large. Choosing the faster algorithm based on p and n yields an upper bound of n^((1 / 2 + o(1)) log n) for p-group isomorphism.
△ Less
Submitted 5 December, 2013;
originally announced December 2013.
-
Bidirectional Collision Detection and Faster Deterministic Isomorphism Testing
Authors:
David J. Rosenbaum
Abstract:
In this work, we introduce bidirectional collision detection --- a new algorithmic tool that applies to the collision problems that arise in many isomorphism problems. For the group isomorphism problem, we show that bidirectional collision detection yields a deterministic n^((1 / 2) log n + O(1)) time algorithm whereas previously the n^(log n + O(1)) generator-enumeration algorithm was the best re…
▽ More
In this work, we introduce bidirectional collision detection --- a new algorithmic tool that applies to the collision problems that arise in many isomorphism problems. For the group isomorphism problem, we show that bidirectional collision detection yields a deterministic n^((1 / 2) log n + O(1)) time algorithm whereas previously the n^(log n + O(1)) generator-enumeration algorithm was the best result for several decades. For the hard special case of solvable groups, we combine bidirectional collision detection with methods from the author's previous work to obtain a deterministic square-root speedup over the best previous algorithm. We also show a deterministic square-root speedup over the best previous algorithm for testing isomorphism of rings. We can even apply bidirectional collision detection to the graph isomorphism problem to obtain a deterministic T^(1 / sqrt(2)) speedup over the best previous deterministic algorithm. Although the space requirements for our algorithms are greater than those for previous deterministic isomorphism tests, we show time-space tradeoffs that interpolate between the resource requirements of our algorithms and previous work.
△ Less
Submitted 16 May, 2013; v1 submitted 14 April, 2013;
originally announced April 2013.
-
Uselessness for an Oracle Model with Internal Randomness
Authors:
Aram W. Harrow,
David J. Rosenbaum
Abstract:
We consider a generalization of the standard oracle model in which the oracle acts on the target with a permutation selected according to internal random coins. We describe several problems that are impossible to solve classically but can be solved by a quantum algorithm using a single query; we show that such infinity-vs-one separations between classical and quantum query complexities can be cons…
▽ More
We consider a generalization of the standard oracle model in which the oracle acts on the target with a permutation selected according to internal random coins. We describe several problems that are impossible to solve classically but can be solved by a quantum algorithm using a single query; we show that such infinity-vs-one separations between classical and quantum query complexities can be constructed from much weaker separations.
We also give conditions to determine when oracle problems---either in the standard model, or in any of the generalizations we consider---cannot be solved with success probability better than random guessing would achieve. In the oracle model with internal randomness where the goal is to gain any nonzero advantage over guessing, we prove (roughly speaking) that $k$ quantum queries are equivalent in power to $2k$ classical queries, thus extending results of Meyer and Pommersheim.
△ Less
Submitted 23 September, 2013; v1 submitted 6 November, 2011;
originally announced November 2011.
