arXiv subjects

Kim-Manuel Klein

Publications and source records attributed to Kim-Manuel Klein.

18 recordsLinked to original sources

arXiv2024

Faster Lattice Basis Computation via a Natural Generalization of the Euclidean Algorithm

The Euclidean algorithm is the oldest algorithms known to mankind. Given two integral numbers $a_1$ and $a_2$, it computes the greatest common divisor (gcd) of $a_1$ and $a_2$ in a very elegant way. From a lattice perspective, it computes a basis of the sum of two one-dimensional lattices $a_1 \mathbb{Z}$ and $a_2 \mathbb{Z}$ as $\gcd(a_1,a_2) \mathbb{Z} = a_1 \mathbb{Z} + a_2 \mathbb{Z}$. In this paper, we show that the classical Euclidean algorithm can be adapted in a very natural way to compute a basis of a general lattice $L(A_1, \ldots , A_n)$ given vectors $A_1, \ldots , A_n \in \mathbb{Z}^d$ with $n> \mathrm{rank}(a_1, \ldots ,a_d)$. Similar to the Euclidean algorithm, our algorithm is very easy to describe and implement and can be written within 12 lines of pseudocode. As our main result, we obtain an algorithm to compute a lattice basis for given vectors $A_1, \ldots , A_n \in \mathbb{Z}^d$ in time (counting bit operations) $LS + \tilde{O}((n-d)d^2 \cdot \log(||A||)$, where $LS$ is the time required to obtain the exact fractional solution of a certain system of linear equalities. The analysis of the running time of our algorithms relies on fundamental statements on the fractionality of solutions of linear systems of equations. So far, the fastest algorithm for lattice basis computation was due to Storjohann and Labhan [SL96] having a running time of $\tilde{O}(nd^\omega\log ||A||)$. For current upper bounds of $LS$, our algorithm has a running time improvement of a factor of at least $d^{0.12}$ over [SL96]. Our algorithm is therefore the first general algorithmic improvement to this classical problem in nearly 30 years. At last, we present a postprocessing procedure which yields an improved size bound of $\sqrt{d} ||A||$ for vectors of the resulting basis matrix.

cs.DS↗

arXiv2024

Tight Lower Bounds for Block-Structured Integer Programs

We study fundamental block-structured integer programs called tree-fold and multi-stage IPs. Tree-fold IPs admit a constraint matrix with independent blocks linked together by few constraints in a recursive pattern; and transposing their constraint matrix yields multi-stage IPs. The state-of-the-art algorithms to solve these IPs have an exponential gap in their running times, making it natural to ask whether this gap is inherent. We answer this question affirmative. Assuming the Exponential Time Hypothesis, we prove lower bounds showing that the exponential difference is necessary, and that the known algorithms are near optimal. Moreover, we prove unconditional lower bounds on the norms of the Graver basis, a fundamental building block of all known algorithms to solve these IPs. This shows that none of the current approaches can be improved beyond this bound.

cs.CC↗

arXiv2023

Simple Lattice Basis Computation -- The Generalization of the Euclidean Algorithm

The Euclidean algorithm is one of the oldest algorithms known to mankind. Given two integral numbers $a_1$ and $a_2$, it computes the greatest common divisor (gcd) of $a_1$ and $a_2$ in a very elegant way. From a lattice perspective, it computes a basis of the sum of two one-dimensional lattices $a_1 \mathbb{Z}$ and $a_2 \mathbb{Z}$ as $\gcd(a_1,a_2) \mathbb{Z} = a_1 \mathbb{Z} + a_2 \mathbb{Z}$. In this paper, we show that the classical Euclidean algorithm can be adapted in a very natural way to compute a basis of a general lattice $L(a_1, \ldots , a_m)$ given vectors $a_1, \ldots , a_m \in \mathbb{Z}^n$ with $m> \mathrm{rank}(a_1, \ldots ,a_m)$. Similar to the Euclidean algorithm, our algorithm is very easy to describe and implement and can be written within 12 lines of pseudocode. While the Euclidean algorithm halves the largest number in every iteration, our generalized algorithm halves the determinant of a full rank subsystem leading to at most $\log (\det B)$ many iterations, for some initial subsystem $B$. Therefore, we can compute a basis of the lattice using at most $\tilde{O}((m-n)n\log(\det B) + mn^{ω-1}\log(||A||_\infty))$ arithmetic operations, where $ω$ is the matrix multiplication exponent and $A = (a_1, \ldots, a_m)$. Even using the worst case Hadamard bound for the determinant, our algorithm improves upon existing algorithm. Another major advantage of our algorithm is that we can bound the entries of the resulting lattice basis by $\tilde{O}(n^2\cdot ||A||_{\infty})$ using a simple pivoting rule. This is in contrast to the typical approach for computing lattice basis, where the Hermite normal form (HNF) is used. In the HNF, entries can be as large as the determinant and hence can only be bounded by an exponential term.

cs.DS↗

arXiv2022

On Minimizing Tardy Processing Time, Max-Min Skewed Convolution, and Triangular Structured ILPs

The starting point of this paper is the problem of scheduling $n$ jobs with processing times and due dates on a single machine so as to minimize the total processing time of tardy jobs, i.e., $1||\sum p_j U_j$. This problem was identified by Bringmann et al. (Algorithmica 2022) as a natural subquadratic-time special case of the classic $1||\sum w_j U_j$ problem, which likely requires time quadratic in the total processing time $P$, because of a fine-grained lower bound. Bringmann et al.~obtain their $\tilde{O}(P^{7/4})$ time scheduling algorithm through a new variant of convolution, dubbed Max-Min Skewed Convolution, which they solve in $\tilde{O}(n^{7/4})$ time. Our main technical contribution is a faster and simpler convolution algorithm running in $\tilde{O}(n^{5/3})$ time. It implies an $\tilde{O}(P^{5/3})$ time algorithm for $1||\sum p_j U_j$, but may also be of independent interest. Inspired by recent developments for the Subset Sum and Knapsack problems, we study $1||\sum p_j U_j$ parameterized by the maximum job processing time $p_{\max}$. With proximity techniques borrowed from integer linear programming (ILP), we show structural properties of the problem that, coupled with a new dynamic programming formulation, lead to an $\tilde{O}(n+p_{\max}^3)$ time algorithm. Moreover, in the setting with multiple machines, we use similar techniques to get an $n \cdot p_{\max}^{O(m)}$ time algorithm for $Pm||\sum p_j U_j$. Finally, we point out that the considered problems exhibit a particular triangular block structure in the constraint matrices of their ILP formulations. In light of recent ILP research, a question that arises is whether one can devise a generic algorithm for such a class of ILPs. We give a negative answer to this question: we show that already a slight generalization of the structure of the scheduling ILP leads to a strongly NP-hard problem.

cs.DS↗

arXiv2022

An Algorithmic Theory of Integer Programming

We study the general integer programming problem where the number of variables $n$ is a variable part of the input. We consider two natural parameters of the constraint matrix $A$: its numeric measure $a$ and its sparsity measure $d$. We show that integer programming can be solved in time $g(a,d)\textrm{poly}(n,L)$, where $g$ is some computable function of the parameters $a$ and $d$, and $L$ is the binary encoding length of the input. In particular, integer programming is fixed-parameter tractable parameterized by $a$ and $d$, and is solvable in polynomial time for every fixed $a$ and $d$. Our results also extend to nonlinear separable convex objective functions. Moreover, for linear objectives, we derive a strongly-polynomial algorithm, that is, with running time $g(a,d)\textrm{poly}(n)$, independent of the rest of the input data. We obtain these results by developing an algorithmic framework based on the idea of iterative augmentation: starting from an initial feasible solution, we show how to quickly find augmenting steps which rapidly converge to an optimum. A central notion in this framework is the Graver basis of the matrix $A$, which constitutes a set of fundamental augmenting steps. The iterative augmentation idea is then enhanced via the use of other techniques such as new and improved bounds on the Graver basis, rapid solution of integer programs with bounded variables, proximity theorems and a new proximity-scaling algorithm, the notion of a reduced objective function, and others. As a consequence of our work, we advance the state of the art of solving block-structured integer programs. In particular, we develop near-linear time algorithms for $n$-fold, tree-fold, and $2$-stage stochastic integer programs. We also discuss some of the many applications of these classes.

math.OC↗

arXiv2021

Collapsing the Tower -- On the Complexity of Multistage Stochastic IPs

In this paper we study the computational complexity of solving a class of block structured integer programs (IPs) - so called multistage stochastic IPs. A multistage stochastic IP is an IP of the form $\max \{ c^T x \mid \mathcal{A} x = b, \,l \leq x \leq u,\, x\text{ integral} \}$ where the constraint matrix $\mathcal{A}$ consists of small block matrices ordered on the diagonal line and for each stage there are larger blocks with few columns connecting the blocks in a tree like fashion. Over the last years there was enormous progress in the area of block structured IPs. For many of the known block IP classes - such as $n$-fold, tree-fold, and two-stage stochastic IPs, nearly matching upper and lower bounds are known concerning their computational complexity. One of the major gaps that remained however was the parameter dependency in the running time for an algorithm solving multistage stochastic IPs. Previous algorithms require a tower of $t$ exponentials, where $t$ is the number of stages, while only a double exponential lower bound was known. In this paper we show that the tower of $t$ exponentials is actually not necessary. We can show an improved running time for the algorithm solving multistage stochastic IPs with a running time of $2^{(d\||A||_\infty)^{\mathcal{O}(d^{3t+1})}} \cdot poly(d,n)$, where $d$ is the sum of columns in the connecting blocks and $n$ is the number of blocks on the lowest stage. In contrast to previous works, our algorithm has only a triple exponential dependency on the parameters and only doubly exponential for every constant $t$. By this we come very close the known double exponential bound (based on the exponential time hypothesis) that holds already for two-stage stochastic IPs, i.e. multistage stochastic IPs with only two stages.

cs.DS↗

arXiv2021

On the Fine-Grained Complexity of the Unbounded SubsetSum and the Frobenius Problem

Consider positive integral solutions $x \in \mathbb{Z}^{n+1}$ to the equation $a_0 x_0 + \ldots + a_n x_n = t$. In the so called unbounded subset sum problem, the objective is to decide whether such a solution exists, whereas in the Frobenius problem, the objective is to compute the largest $t$ such that there is no such solution. In this paper we study the algorithmic complexity of the unbounded subset sum, the Frobenius problem and a generalization of the problems. More precisely, we study pseudo-polynomial time algorithms with a running time that depends on the smallest number $a_0$ or respectively the largest number $a_n$. For the parameter $a_0$, we show that all considered problems are subquadratically equivalent to $(min,+)$-convolution, a fundamental algorithmic problem from the area of fine-grained complexity. By this equivalence, we obtain hardness results for the considered problems (based on the assumption that an algorithm with a subquadratic running time for $(min,+)$-convolution does not exist) as well as algorithms with improved running time. The proof for the equivalence makes use of structural properties of solutions, a technique that was developed in the area of integer programming. In case of the complexity of the problems parameterized by $a_n$, we present improved algorithms. For example we give a quasi linear time algorithm for the Frobenius problem as well as a hardness result based on the strong exponential time hypothesis.

cs.DS↗

arXiv2021

The Double Exponential Runtime is Tight for 2-Stage Stochastic ILPs

We consider fundamental algorithmic number theoretic problems and their relation to a class of block structured Integer Linear Programs (ILPs) called $2$-stage stochastic. A $2$-stage stochastic ILP is an integer program of the form $\min \{c^T x \mid \mathcal{A} x = b, \ell \leq x \leq u, x \in \mathbb{Z}^{r + ns} \}$ where the constraint matrix $\mathcal{A} \in \mathbb{Z}^{nt \times r +ns}$ consists of $n$ matrices $A_i \in \mathbb{Z}^{t \times r}$ on the vertical line and $n$ matrices $B_i \in \mathbb{Z}^{t \times s}$ on the diagonal line aside. First, we show a stronger hardness result for a number theoretic problem called Quadratic Congruences where the objective is to compute a number $z \leq γ$ satisfying $z^2 \equiv α\bmod β$ for given $α, β, γ\in \mathbb{Z}$. This problem was proven to be NP-hard already in 1978 by Manders and Adleman. However, this hardness only applies for instances where the prime factorization of $β$ admits large multiplicities of each prime number. We circumvent this necessity proving that the problem remains NP-hard, even if each prime number only occurs constantly often. Then, using this new hardness result for the Quadratic Congruences problem, we prove a lower bound of $2^{2^{δ(s+t)}} |I|^{O(1)}$ for some $δ> 0$ for the running time of any algorithm solving $2$-stage stochastic ILPs assuming the Exponential Time Hypothesis (ETH). Here, $|I|$ is the encoding length of the instance. This result even holds if $r$, $||b||_{\infty}$, $||c||_{\infty}, ||\ell||_{\infty}$ and the largest absolute value $Δ$ in the constraint matrix $\mathcal{A}$ are constant. This shows that the state-of-the-art algorithms are nearly tight. Further, it proves the suspicion that these ILPs are indeed harder to solve than the closely related $n$-fold ILPs where the contraint matrix is the transpose of $\mathcal A$.

cs.CC↗

arXiv2020

Fuzzy Simultaneous Congruences

We introduce a very natural generalization of the well-known problem of simultaneous congruences. Instead of searching for a positive integer $s$ that is specified by $n$ fixed remainders modulo integer divisors $a_1,\dots,a_n$ we consider remainder intervals $R_1,\dots,R_n$ such that $s$ is feasible if and only if $s$ is congruent to $r_i$ modulo $a_i$ for some remainder $r_i$ in interval $R_i$ for all $i$. This problem is a special case of a 2-stage integer program with only two variables per constraint which is is closely related to directed Diophantine approximation as well as the mixing set problem. We give a hardness result showing that the problem is NP-hard in general. By investigating the case of harmonic divisors, i.e. $a_{i+1}/a_i$ is an integer for all $i<n$, which was heavily studied for the mixing set problem as well, we also answer a recent algorithmic question from the field of real-time systems. We present an algorithm to decide the feasibility of an instance in time $\mathcal{O}(n^2)$ and we show that if it exists even the smallest feasible solution can be computed in strongly polynomial time $\mathcal{O}(n^3)$.

cs.DM↗

arXiv2020

New Bounds for the Vertices of the Integer Hull

The vertices of the integer hull are the integral equivalent to the well-studied basic feasible solutions of linear programs. In this paper we give new bounds on the number of non-zero components -- their support -- of these vertices matching either the best known bounds or improving upon them. While the best known bounds make use of deep techniques, we only use basic results from probability theory to make use of the concentration of measure effect. To show the versatility of our techniques, we use our results to give the best known bounds on the number of such vertices and an algorithm to enumerate them. We also improve upon the known lower bounds to show that our results are nearly optimal. One of the main ingredients of our work is a generalization of the famous Hoeffding bound to vector-valued random variables that might be of general interest.

cs.DS↗

arXiv2019

About the Complexity of Two-Stage Stochastic IPs

We consider so called $2$-stage stochastic integer programs (IPs) and their generalized form of multi-stage stochastic IPs. A $2$-stage stochastic IP is an integer program of the form $\max \{ c^T x \mid Ax = b, l \leq x \leq u, x \in \mathbb{Z}^{nt + s} \}$ where the constraint matrix $A \in \mathbb{Z}^{r \times s}$ consists roughly of $n$ repetition of a block matrix $A$ on the vertical line and $n$ repetitions of a matrix $B \in \mathbb{Z}^{r \times t}$ on the diagonal. In this paper we improve upon an algorithmic result by Hemmecke and Schultz form 2003 to solve $2$-stage stochastic IPs. The algorithm is based on the Graver augmentation framework where our main contribution is to give an explicit doubly exponential bound on the size of the augmenting steps. The previous bound for the size of the augmenting steps relied on non-constructive finiteness arguments from commutative algebra and therefore only an implicit bound was known that depends on parameters $r,s,t$ and $Δ$, where $Δ$ is the largest entry of the constraint matrix. Our new improved bound however is obtained by a novel theorem which argues about the intersection of paths in a vector space. As a result of our new bound we obtain an algorithm to solve $2$-stage stochastic IPs in time $poly(n,t) \cdot f(r,s,Δ)$, where $f$ is a doubly exponential function. To complement our result, we also prove a doubly exponential lower bound for the size of the augmenting steps.

cs.DS↗

arXiv2018

Empowering the Configuration-IP $-$ New PTAS Results for Scheduling with Setups Times

Integer linear programs of configurations, or configuration IPs, are a classical tool in the design of algorithms for scheduling and packing problems, where a set of items has to be placed in multiple target locations. Herein a configuration describes a possible placement on one of the target locations, and the IP is used to chose suitable configurations covering the items. We give an augmented IP formulation, which we call the module configuration IP. It can be described within the framework of n-fold integer programming and therefore be solved efficiently. As an application, we consider scheduling problems with setup times, in which a set of jobs has to be scheduled on a set of identical machines, with the objective of minimizing the makespan. For instance, we investigate the case that jobs can be split and scheduled on multiple machines. However, before a part of a job can be processed an uninterrupted setup depending on the job has to be paid. For both of the variants that jobs can be executed in parallel or not, we obtain an efficient polynomial time approximation scheme (EPTAS) of running time $f(1/\varepsilon)\times \mathrm{poly}(|I|)$ with a single exponential term in $f$ for the first and a double exponential one for the second case. Previously, only constant factor approximations of $5/3$ and $4/3 + \varepsilon$ respectively were known. Furthermore, we present an EPTAS for a problem where classes of (non-splittable) jobs are given, and a setup has to be paid for each class of jobs being executed on one machine.

cs.DS↗

arXiv2018

Online Strip Packing with Polynomial Migration

We consider the relaxed online strip packing problem: Rectangular items arrive online and have to be packed without rotations into a strip of fixed width such that the packing height is minimized. Thereby, repacking of previously packed items is allowed. The amount of repacking is measured by the migration factor, defined as the total size of repacked items divided by the size of the arriving item. First, we show that no algorithm with constant migration factor can produce solutions with asymptotic ratio better than 4/3. Against this background, we allow amortized migration, i.e. to save migration for a later time step. As a main result, we present an AFPTAS with asymptotic ratio $1 + \mathcal{O}(ε)$ for any $ε> 0$ and amortized migration factor polynomial in $1 / ε$. To our best knowledge, this is the first algorithm for online strip packing considered in a repacking model.

cs.DS↗

arXiv2018

Faster Algorithms for Integer Programs with Block Structure

We consider integer programming problems $\max \{ c^T x : \mathcal{A} x = b, l \leq x \leq u, x \in \mathbb{Z}^{nt}\}$ where $\mathcal{A}$ has a (recursive) block-structure generalizing "$n$-fold integer programs" which recently received considerable attention in the literature. An $n$-fold IP is an integer program where $\mathcal{A}$ consists of $n$ repetitions of submatrices $A \in \mathbb{Z}^{r \times t}$ on the top horizontal part and $n$ repetitions of a matrix $B \in \mathbb{Z}^{s \times t}$ on the diagonal below the top part. Instead of allowing only two types of block matrices, one for the horizontal line and one for the diagonal, we generalize the $n$-fold setting to allow for arbitrary matrices in every block. We show that such an integer program can be solved in time $n^2 t^2 ϕ \cdot (rsΔ)^{\mathcal{O}(rs^2+ sr^2)}$ (ignoring logarithmic factors). Here $Δ$ is an upper bound on the largest absolute value of an entry of $\mathcal{A}$ and $ϕ$ is the largest binary encoding length of a coefficient of $c$. This improves upon the previously best algorithm of Hemmecke, Onn and Romanchuk that runs in time $n^3t^3 ϕ \cdot Δ^{\mathcal{O}(t^2s)}$. In particular, our algorithm is not exponential in the number $t$ of columns of $A$ and $B$. Our algorithm is based on a new upper bound on the $l_1$-norm of an element of the "Graver basis" of an integer matrix and on a proximity bound between the LP and IP optimal solutions tailored for IPs with block structure. These new bounds rely on the "Steinitz Lemma". Furthermore, we extend our techniques to the recently introduced "tree-fold IPs", where we again present a more efficient algorithm in a generalized setting.

cs.DM↗

arXiv2016

About the Structure of the Integer Cone and its Application to Bin Packing

We consider the bin packing problem with $d$ different item sizes and revisit the structure theorem given by Goemans and Rothvoß[6] about solutions of the integer cone. We present new techniques on how solutions can be modified and give a new structure theorem that relies on the set of vertices of the underlying integer polytope. As a result of our new structure theorem, we obtain an algorithm for the bin packing problem with running time $|V|^{2^{O(d)}} \cdot enc(I)^{O(1)}$, where $V$ is the set of vertices of the integer knapsack polytope and $enc(I)$ is the encoding length of the bin packing instance. The algorithm is fixed parameter tractable, parameterized by the number of vertices of the integer knapsack polytope $|V|$. This shows that the bin packing problem can be solved efficiently when the underlying integer knapsack polytope has an easy structure, i.e. has a small number of vertices. Furthermore, we show that the presented bounds of the structure theorem are asymptotically tight. We give a construction of bin packing instances using new structural insights and classical number theoretical theorems which yield the desired lower bound.

cs.DS↗

arXiv2016

Closing the Gap for Makespan Scheduling via Sparsification Techniques

Makespan scheduling on identical machines is one of the most basic and fundamental packing problems studied in the discrete optimization literature. It asks for an assignment of $n$ jobs to a set of $m$ identical machines that minimizes the makespan. The problem is strongly NP-hard, and thus we do not expect a $(1+ε)$-approximation algorithm with a running time that depends polynomially on $1/ε$. Furthermore, Chen et al. [3] recently showed that a running time of $2^{(1/ε)^{1-δ}}+\text{poly}(n)$ for any $δ>0$ would imply that the Exponential Time Hypothesis (ETH) fails. A long sequence of algorithms have been developed that try to obtain low dependencies on $1/ε$, the better of which achieves a running time of $2^{\tilde{O}(1/ε^2)}+O(n\log n)$ [11]. In this paper we obtain an algorithm with a running time of $2^{\tilde{O}(1/ε)}+O(n\log n)$, which is tight under ETH up to logarithmic factors on the exponent. Our main technical contribution is a new structural result on the configuration-IP. More precisely, we show the existence of a highly symmetric and sparse optimal solution, in which all but a constant number of machines are assigned a configuration with small support. This structure can then be exploited by integer programming techniques and enumeration. We believe that our structural result is of independent interest and should find applications to other settings. In particular, we show how the structure can be applied to the minimum makespan problem on related machines and to a larger class of objective functions on parallel machines. For all these cases we obtain an efficient PTAS with running time $2^{\tilde{O}(1/ε)} + \text{poly}(n)$.

cs.DS↗

arXiv2015

Fully Dynamic Bin Packing Revisited

We consider the fully dynamic bin packing problem, where items arrive and depart in an online fashion and repacking of previously packed items is allowed. The goal is, of course, to minimize both the number of bins used as well as the amount of repacking. A recently introduced way of measuring the repacking costs at each timestep is the migration factor, defined as the total size of repacked items divided by the size of an arriving or departing item. Concerning the trade-off between number of bins and migration factor, if we wish to achieve an asymptotic competitive ration of $1 + ε$ for the number of bins, a relatively simple argument proves a lower bound of $Ω(\frac{1}ε)$ for the migration factor. We establish a nearly matching upper bound of $O(\frac{1}ε^4 \log \frac{1}ε)$ using a new dynamic rounding technique and new ideas to handle small items in a dynamic setting such that no amortization is needed. The running time of our algorithm is polynomial in the number of items $n$ and in $\frac{1}ε$. The previous best trade-off was for an asymptotic competitive ratio of $\frac{5}{4}$ for the bins (rather than $1+ε$) and needed an amortized number of $O(\log n)$ repackings (while in our scheme the number of repackings is independent of $n$ and non-amortized).

cs.DS↗

arXiv2013

A Robust AFPTAS for Online Bin Packing with Polynomial Migration

In this paper we develop general LP and ILP techniques to find an approximate solution with improved objective value close to an existing solution. The task of improving an approximate solution is closely related to a classical theorem of Cook et al. in the sensitivity analysis for LPs and ILPs. This result is often applied in designing robust algorithms for online problems. We apply our new techniques to the online bin packing problem, where it is allowed to reassign a certain number of items, measured by the migration factor. The migration factor is defined by the total size of reassigned items divided by the size of the arriving item. We obtain a robust asymptotic fully polynomial time approximation scheme (AFPTAS) for the online bin packing problem with migration factor bounded by a polynomial in $\frac{1}ε$. This answers an open question stated by Epstein and Levin in the affirmative. As a byproduct we prove an approximate variant of the sensitivity theorem by Cook at el. for linear programs.

cs.DS↗