亚洲男人的天堂2018av,欧美草比,久久久久久免费视频精选,国色天香在线看免费,久久久久亚洲av成人片仓井空

Min-plus product of two $n\times n$ matrices is a fundamental problem in algorithm research. It is known to be equivalent to APSP, and in general it has no truly subcubic algorithms. In this paper, we focus on the min-plus product on a special class of matrices, called $\delta$-bounded-difference matrices, in which the difference between any two adjacent entries is bounded by $\delta=O(1)$. Our algorithm runs in randomized time $O(n^{2.779})$ by the fast rectangular matrix multiplication algorithm [Le Gall \& Urrutia 18], better than $\tilde{O}(n^{2+\omega/3})=O(n^{2.791})$ ($\omega<2.373$ [Alman \& V.V.Williams 20]). This improves previous result of $\tilde{O}(n^{2.824})$ [Bringmann et al. 16]. When $\omega=2$ in the ideal case, our complexity is $\tilde{O}(n^{2+2/3})$, improving Bringmann et al.'s result of $\tilde{O}(n^{2.755})$.

We consider a family of unadjusted HMC samplers, which includes standard position HMC samplers and discretizations of the underdamped Langevin process. A detailed analysis and optimization of the parameters is conducted in the Gaussian case. Then, a stochastic gradient version of the samplers is considered, for which dimension-free convergence rates are established for log-concave smooth targets, gathering in a unified framework previous results on both processes. Both results indicate that partial refreshments of the velocity are more efficient than standard full refreshments.

Softmax policy gradient is a popular algorithm for policy optimization in single-agent reinforcement learning, particularly since projection is not needed for each gradient update. However, in multi-agent systems, the lack of central coordination introduces significant additional difficulties in the convergence analysis. Even for a stochastic game with identical interest, there can be multiple Nash Equilibria (NEs), which disables proof techniques that rely on the existence of a unique global optimum. Moreover, the softmax parameterization introduces non-NE policies with zero gradient, making NE-seeking difficult for gradient-based algorithms. In this paper, we study the finite time convergence of decentralized softmax gradient play in a special form of game, Markov Potential Games (MPGs), which includes the identical interest game as a special case. We investigate both gradient play and natural gradient play, with and without $\log$-barrier regularization. Establishing convergence for the unregularized cases relies on an assumption that the stationary policies are isolated, and yields convergence bounds that contain a trajectory dependent constant that can be arbitrarily large. We introduce the $\log$-barrier regularization to overcome these drawbacks, with the cost of slightly worse dependence on other factors such as the action set size. An empirical study on an identical interest matrix game confirms the theoretical findings.

In this paper, we study the interference cancellation capabilities of receivers and transmitters in multiple-input-multiple-output (MIMO) systems using theoretical calculations and numerical simulations in Quadriga. We study so-called Reduced Channel Zero-Forcing (RCZF) class of precoding as well as Minimum MSE Interference Rejection Combiner (MMSE-IRC) and QR Maximum Likelihood Detection (QR-MLD) receivers. Based on very simple but extremely useful algebraic manipulations, their asymptotical equivalence is proven analytically and demonstrated via simulations. Our theoretical and experimental results confirm that MMSE-IRC and QR-MLD receivers in combination with the RCZF precoding provide complete interference suppression asymptotically.

In this paper, we focus on the construction of a hybrid scheme for the approximation of non-Maxwellian kinetic models with uncertainties. In the context of multiagent systems, the introduction of a kernel at the kinetic level is useful to avoid unphysical interactions. The methods here proposed, combine a direct simulation Monte Carlo (DSMC) in the phase space together with stochastic Galerkin (sG) methods in the random space. The developed schemes preserve the main physical properties of the solution together with accuracy in the random space. The consistency of the methods is tested with respect to surrogate Fokker-Planck models that can be obtained in the quasi-invariant regime of parameters. Several applications of the schemes to non-Maxwellian models of multiagent systems are reported.

In this paper, we show several parameterized problems to be complete for the class XNLP: parameterized problems that can be solved with a non-deterministic algorithm that uses $f(k)\log n$ space and $f(k)n^c$ time, with $f$ a computable function, $n$ the input size, $k$ the parameter and $c$ a constant. The problems include Maximum Regular Induced Subgraph and Max Cut parameterized by linear clique-width, Capacitated (Red-Blue) Dominating Set parameterized by pathwidth, Odd Cycle Transversal parameterized by a new parameter we call logarithmic linear clique-width (defined as $k/\log n$ for an $n$-vertex graph of linear clique-width $k$), and Bipartite Bandwidth.

We study the problem of computing the vitality with respect to max flow of edges and vertices in undirected planar graphs, where the vitality of an edge/vertex in a graph with respect to max flow between two fixed vertices $s,t$ is defined as the max flow decrease when the edge/vertex is removed from the graph. We show that the vitality of any $k$ selected edges can be computed in $O(kn + n\log\log n)$ worst-case time, and that a $\delta$ additive approximation of the vitality of all edges with capacity at most $c$ can be computed in $O(\frac{c}{\delta}n +n \log \log n)$ worst-case time, where $n$ is the size of the graph. Similar results are given for the vitality of vertices. All our algorithms work in $O(n)$ space.

We propose throughput and cost optimal job scheduling algorithms in cloud computing platforms offering Infrastructure as a Service. We first consider online migration and propose job scheduling algorithms to minimize job migration and server running costs. We consider algorithms that assume knowledge of job-size on arrival of jobs. We characterize the optimal cost subject to system stability. We develop a drift-plus-penalty framework based algorithm that can achieve optimal cost arbitrarily closely. Specifically this algorithm yields a trade-off between delay and costs. We then relax the job-size knowledge assumption and give an algorithm that uses readily offered service to the jobs. We show that this algorithm gives order-wise identical cost as the job size based algorithm. Later, we consider offline job migration that incurs migration delays. We again present throughput optimal algorithms that minimize server running cost. We illustrate the performance of the proposed algorithms and compare these to the existing algorithms via simulation.

Aligning a sequence to a walk in a labeled graph is a problem of fundamental importance to Computational Biology. For finding a walk in an arbitrary graph with $|E|$ edges that exactly matches a pattern of length $m$, a lower bound based on the Strong Exponential Time Hypothesis (SETH) implies an algorithm significantly faster than $O(|E|m)$ time is unlikely [Equi et al., ICALP 2019]. However, for many special graphs, such as de Bruijn graphs, the problem can be solved in linear time [Bowe et al., WABI 2012]. For approximate matching, the picture is more complex. When edits (substitutions, insertions, and deletions) are only allowed to the pattern, or when the graph is acyclic, the problem is again solvable in $O(|E|m)$ time. When edits are allowed to arbitrary cyclic graphs, the problem becomes NP-complete, even on binary alphabets [Jain et al., RECOMB 2019]. These results hold even when edits are restricted to only substitutions. The complexity of approximate pattern matching on de Bruijn graphs remained open. We investigate this problem and show that the properties that make de Bruijn graphs amenable to efficient exact pattern matching do not extend to approximate matching, even when restricted to the substitutions only case with alphabet size four. We prove that determining the existence of a matching walk in a de Bruijn graph is NP-complete when substitutions are allowed to the graph. In addition, we demonstrate that an algorithm significantly faster than $O(|E|m)$ is unlikely for de Bruijn graphs in the case where only substitutions are allowed to the pattern. This stands in contrast to pattern-to-text matching where exact matching is solvable in linear time, like on de Bruijn graphs, but approximate matching under substitutions is solvable in subquadratic $O(n\sqrt{m})$ time, where $n$ is the text's length [Abrahamson, SIAM J. Computing 1987].

In this paper, high-order numerical integrators on homogeneous spaces will be presented as an application of nonholonomic partitioned Runge-Kutta Munthe-Kaas (RKMK) methods on Lie groups. A homogeneous space $M$ is a manifold where a group $G$ acts transitively. Such a space can be understood as a quotient $M \cong G/H$, where $H$ a closed Lie subgroup, is the isotropy group of each point of $M$. The Lie algebra of $G$ may be decomposed into $\mathfrak{g} = \mathfrak{m} \oplus \mathfrak{h}$, where $\mathfrak{h}$ is the subalgebra that generates $H$ and $\mathfrak{m}$ is a subspace. Thus, variational problems on $M$ can be treated as nonholonomically constrained problems on $G$, by requiring variations to remain on $\mathfrak{m}$. Nonholonomic partitioned RKMK integrators are derived as a modification of those obtained by a discrete variational principle on Lie groups, and can be interpreted as obeying a discrete Chetaev principle. These integrators tend to preserve several properties of their purely variational counterparts.