arXiv subjects

Anupam Gupta

Publications and source records attributed to Anupam Gupta.

At least 109 records · Page 6Linked to original sources

arXiv2018

A Nearly-Linear Bound for Chasing Nested Convex Bodies

Friedman and Linial introduced the convex body chasing problem to explore the interplay between geometry and competitive ratio in metrical task systems. In convex body chasing, at each time step $t \in \mathbb{N}$, the online algorithm receives a request in the form of a convex body $K_t \subseteq \mathbb{R}^d$ and must output a point $x_t \in K_t$. The goal is to minimize the total movement between consecutive output points, where the distance is measured in some given norm. This problem is still far from being understood, and recently Bansal et al. gave an algorithm for the nested version, where each convex body is contained within the previous one. We propose a different strategy which is $O(d \log d)$-competitive algorithm for this nested convex body chasing problem, improving substantially over previous work. Our algorithm works for any norm. This result is almost tight, given an $Ω(d)$ lower bound for the $\ell_{\infty}$.

cs.DS↗

arXiv2018

Losing Treewidth by Separating Subsets

We study the problem of deleting the smallest set $S$ of vertices (resp. edges) from a given graph $G$ such that the induced subgraph (resp. subgraph) $G \setminus S$ belongs to some class $\mathcal{H}$. We consider the case where graphs in $\mathcal{H}$ have treewidth bounded by $t$, and give a general framework to obtain approximation algorithms for both vertex and edge-deletion settings from approximation algorithms for certain natural graph partitioning problems called $k$-Subset Vertex Separator and $k$-Subset Edge Separator, respectively. For the vertex deletion setting, our framework combined with the current best result for $k$-Subset Vertex Separator, yields a significant improvement in the approximation ratios for basic problems such as $k$-Treewidth Vertex Deletion and Planar-$F$ Vertex Deletion. Our algorithms are simpler than previous works and give the first uniform approximation algorithms under the natural parameterization. For the edge deletion setting, we give improved approximation algorithms for $k$-Subset Edge Separator combining ideas from LP relaxations and important separators. We present their applications in bounded-degree graphs, and also give an APX-hardness result for the edge deletion problems.

cs.DS↗

arXiv2018

Conditional stability of particle alignment in finite-Reynolds-number channel flow

Finite-size neutrally buoyant particles in a channel flow are known to accumulate at specific equilibrium positions or spots in the channel cross-section if the flow inertia is finite at the particle scale. Experiments in different conduit geometries have shown that while reaching equilibrium locations, particles tend also to align regularly in the streamwise direction. In this paper, the Force Coupling Method was used to numerically investigate the inertia-induced particle alignment, using square channel geometry. The method was first shown to be suitable to capture the quasi-steady lift force that leads to particle cross-streamline migration in channel flow. Then the particle alignment in the flow direction was investigated by calculating the particle relative trajectories as a function of flow inertia and of the ratio between the particle size and channel hydraulic diameter. The flow streamlines were examined around the freely rotating particles at equilibrium, revealing stable small-scale vortices between aligned particles. The streamwise inter-particle spacing between aligned particles at equilibrium was calculated and compared to available experimental data in square channel flow (Gao {\it et al.} Microfluidics and Nanofluidics {\bf 21}, 154 (2017)). The new result highlighted by our numerical simulations is that the inter-particle spacing is unconditionally stable only for a limited number of aligned particles in a single train, the threshold number being dependent on the confinement (particle-to-channel size ratio) and on the Reynolds number. For instance, when the particle Reynolds number is $\approx1$ and the particle-to-channel height size ratio is $\approx0.1$, the maximum number of stable aligned particles per train is equal to 3. This agrees with statistics realized on the experiments of (Gao {\it et al.} Microfluidics and Nanofluidics {\bf 21}, 154 (2017)).

physics.flu-dyn↗

arXiv2018

Inertial spheroids in homogeneous, isotropic turbulence

We study the rotational dynamics of {\it inertial} disks and rods in three-dimensional, homogeneous isotropic turbulence. In particular, we show how the alignment and the decorrelation time-scales of such spheroids depend, critically, on both the level of inertia and the aspect ratio of these particles. These results illustrate the effect of inertia---which leads to a preferential sampling of the local flow geometry---on the statistics of both disks and rods in a turbulent flow. Our results are important for a variety of natural and industrial settings where the turbulent transport of asymmetric, spheroidal inertial particles is ubiquitous.

physics.flu-dyn↗

arXiv2018

Non-Preemptive Flow-Time Minimization via Rejections

We consider the online problem of minimizing weighted flow-time on unrelated machines. Although much is known about this problem in the resource-augmentation setting, these results assume that jobs can be preempted. We give the first constant-competitive algorithm for the non-preemptive setting in the rejection model. In this rejection model, we are allowed to reject an $\varepsilon$-fraction of the total weight of jobs, and compare the resulting flow-time to that of the offline optimum which is required to schedule all jobs. This is arguably the weakest assumption in which such a result is known for weighted flow-time on unrelated machines. While our algorithms are simple, we need a delicate dual-fitting argument to bound the flow-time while only a small fraction of elements are rejected.

cs.DS↗

arXiv2018

Fully-Dynamic Bin Packing with Limited Repacking

We study the classic Bin Packing problem in a fully-dynamic setting, where new items can arrive and old items may depart. We want algorithms with low asymptotic competitive ratio \emph{while repacking items sparingly} between updates. Formally, each item $i$ has a \emph{movement cost} $c_i\geq 0$, and we want to use $α\cdot OPT$ bins and incur a movement cost $γ\cdot c_i$, either in the worst case, or in an amortized sense, for $α, γ$ as small as possible. We call $γ$ the \emph{recourse} of the algorithm. This is motivated by cloud storage applications, where fully-dynamic Bin Packing models the problem of data backup to minimize the number of disks used, as well as communication incurred in moving file backups between disks. Since the set of files changes over time, we could recompute a solution periodically from scratch, but this would give a high number of disk rewrites, incurring a high energy cost and possible wear and tear of the disks. In this work, we present optimal tradeoffs between number of bins used and number of items repacked, as well as natural extensions of the latter measure.

cs.DS↗

arXiv2018

Maximizing Profit with Convex Costs in the Random-order Model

Suppose a set of requests arrives online: each request gives some value $v_i$ if accepted, but requires using some amount of each of $d$ resources. Our cost is a convex function of the vector of total utilization of these $d$ resources. Which requests should be accept to maximize our profit, i.e., the sum of values of the accepted demands, minus the convex cost? We consider this problem in the random-order a.k.a. secretary model, and show an $O(d)$-competitive algorithm for the case where the convex cost function is also supermodular. If the set of accepted demands must also be independent in a given matroid, we give an $O(d^3 α)$-competitive algorithm for the supermodular case, and an improved $O(d^2α)$ if the convex cost function is also separable. Here $α$ is the competitive ratio of the best algorithm for the submodular secretary problem. These extend and improve previous results known for this problem. Our techniques are simple but use powerful ideas from convex duality, which give clean interpretations of existing work, and allow us to give the extensions and improvements.

cs.DS↗

arXiv2018

Droplet breakup driven by shear thinning solutions in a microfluidic T-Junction

Droplet-based microfluidics turned out to be an efficient and adjustable platform for digital analysis, encapsulation of cells, drug formulation, and polymerase chain reaction. Typically, for most biomedical applications, the handling of complex, non-Newtonian fluids is involved, e.g. synovial and salivary fluids, collagen, and gel scaffolds. In this study we investigate the problem of droplet formation occurring in a microfluidic T-shaped junction, when the continuous phase is made of shear thinning liquids. At first, we review in detail the breakup process providing extensive, side-by-side comparisons between Newtonian and non-Newtonian liquids over unexplored ranges of flow conditions and viscous responses. The non-Newtonian liquid carrying the droplets is made of Xanthan solutions, a stiff rod-like polysaccharide displaying a marked shear thinning rheology. By defining an effective Capillary number, a simple yet effective methodology is used to account for the shear-dependent viscous response occurring at the breakup. The droplet size can be predicted over a wide range of flow conditions simply by knowing the rheology of the bulk continuous phase. Experimental results are complemented with numerical simulations of purely shear thinning fluids using Lattice Boltzmann models. The good agreement between the experimental and numerical data confirm the validity of the proposed rescaling with the effective Capillary number.

cond-mat.soft↗

arXiv2017

Heavy inertial particles in turbulent flows gain energy slowly but lose it rapidly

We present an extensive numerical study of the time irreversibility of the dynamics of heavy inertial particles in three-dimensional, statistically homogeneous and isotropic turbulent flows. We show that the probability density function (PDF) of the increment, $W(τ)$, of a particle's energy over a time-scale $τ$ is non-Gaussian, and skewed towards negative values. This implies that, on average, particles gain energy over a period of time that is longer than the duration over which they lose energy. We call this $\textit{slow gain}$ and $\textit{fast loss}$. We find that the third moment of $W(τ)$ scales as $τ^3$, for small values of $τ$. We show that the PDF of power-input $p$ is negatively skewed too; we use this skewness ${\rm Ir}$ as a measure of the time-irreversibility and we demonstrate that it increases sharply with the Stokes number ${\rm St}$, for small ${\rm St}$; this increase slows down at ${\rm St} \simeq 1$. Furthermore, we obtain the PDFs of $t^+$ and $t^-$, the times over which $p$ has, respectively, positive or negative signs, i.e., the particle gains or loses energy. We obtain from these PDFs a direct and natural quantification of the the slow-gain and fast-loss of the particles, because these PDFs possess exponential tails, whence we infer the characteristic loss and gain times $t_{\rm loss}$ and $t_{\rm gain}$, respectively; and we obtain $t_{\rm loss} < t_{\rm gain}$, for all the cases we have considered. Finally, we show that the slow-gain in energy of the particles is equally likely in vortical or strain-dominated regions of the flow; in contrast, the fast-loss of energy occurs with greater probability in the latter than in the former.

physics.flu-dyn↗

arXiv2017

The role of BKM-type theorems in $3D$ Euler, Navier-Stokes and Cahn-Hilliard-Navier-Stokes analysis

The Beale-Kato-Majda theorem contains a single criterion that controls the behaviour of solutions of the $3D$ incompressible Euler equations. Versions of this theorem are discussed in terms of the regularity issues surrounding the $3D$ incompressible Euler and Navier-Stokes equations together with a phase-field model for the statistical mechanics of binary mixtures called the $3D$ Cahn-Hilliard-Navier-Stokes (CHNS) equations. A theorem of BKM-type is established for the CHNS equations for the full parameter range. Moreover, for this latter set, it is shown that there exists a Reynolds number and a bound on the energy-dissipation rate that, remarkably, reproduces the $Re^{3/4}$ upper bound on the inverse Kolmogorov length normally associated with the Navier-Stokes equations alone. An alternative length-scale is introduced and discussed, together with a set of pseudo-spectral computations on a $128^{3}$ grid.

nlin.CD↗

arXiv2017

An FPT Algorithm Beating 2-Approximation for $k$-Cut

In the $k$-Cut problem, we are given an edge-weighted graph $G$ and an integer $k$, and have to remove a set of edges with minimum total weight so that $G$ has at least $k$ connected components. Prior work on this problem gives, for all $h \in [2,k]$, a $(2-h/k)$-approximation algorithm for $k$-cut that runs in time $n^{O(h)}$. Hence to get a $(2 - \varepsilon)$-approximation algorithm for some absolute constant $\varepsilon$, the best runtime using prior techniques is $n^{O(k\varepsilon)}$. Moreover, it was recently shown that getting a $(2 - \varepsilon)$-approximation for general $k$ is NP-hard, assuming the Small Set Expansion Hypothesis. If we use the size of the cut as the parameter, an FPT algorithm to find the exact $k$-Cut is known, but solving the $k$-Cut problem exactly is $W[1]$-hard if we parameterize only by the natural parameter of $k$. An immediate question is: \emph{can we approximate $k$-Cut better in FPT-time, using $k$ as the parameter?} We answer this question positively. We show that for some absolute constant $\varepsilon > 0$, there exists a $(2 - \varepsilon)$-approximation algorithm that runs in time $2^{O(k^6)} \cdot \widetilde{O} (n^4)$. This is the first FPT algorithm that is parameterized only by $k$ and strictly improves the $2$-approximation.

cs.DS↗

arXiv2017

A Local-Search Algorithm for Steiner Forest

In the Steiner Forest problem, we are given a graph and a collection of source-sink pairs, and the goal is to find a subgraph of minimum total length such that all pairs are connected. The problem is APX-Hard and can be 2-approximated by, e.g., the elegant primal-dual algorithm of Agrawal, Klein, and Ravi from 1995. We give a local-search-based constant-factor approximation for the problem. Local search brings in new techniques to an area that has for long not seen any improvements and might be a step towards a combinatorial algorithm for the more general survivable network design problem. Moreover, local search was an essential tool to tackle the dynamic MST/Steiner Tree problem, whereas dynamic Steiner Forest is still wide open. It is easy to see that any constant factor local search algorithm requires steps that add/drop many edges together. We propose natural local moves which, at each step, either (a) add a shortest path in the current graph and then drop a bunch of inessential edges, or (b) add a set of edges to the current solution. This second type of moves is motivated by the potential function we use to measure progress, combining the cost of the solution with a penalty for each connected component. Our carefully-chosen local moves and potential function work in tandem to eliminate bad local minima that arise when using more traditional local moves.

cs.DS↗

arXiv2017

Nearly Optimal Sampling Algorithms for Combinatorial Pure Exploration

We study the combinatorial pure exploration problem Best-Set in stochastic multi-armed bandits. In a Best-Set instance, we are given $n$ arms with unknown reward distributions, as well as a family $\mathcal{F}$ of feasible subsets over the arms. Our goal is to identify the feasible subset in $\mathcal{F}$ with the maximum total mean using as few samples as possible. The problem generalizes the classical best arm identification problem and the top-$k$ arm identification problem, both of which have attracted significant attention in recent years. We provide a novel instance-wise lower bound for the sample complexity of the problem, as well as a nontrivial sampling algorithm, matching the lower bound up to a factor of $\ln|\mathcal{F}|$. For an important class of combinatorial families, we also provide polynomial time implementation of the sampling algorithm, using the equivalence of separation and optimization for convex program, and approximate Pareto curves in multi-objective optimization. We also show that the $\ln|\mathcal{F}|$ factor is inevitable in general through a nontrivial lower bound construction. Our results significantly improve several previous results for several important combinatorial constraints, and provide a tighter understanding of the general Best-Set problem. We further introduce an even more general problem, formulated in geometric terms. We are given $n$ Gaussian arms with unknown means and unit variance. Consider the $n$-dimensional Euclidean space $\mathbb{R}^n$, and a collection $\mathcal{O}$ of disjoint subsets. Our goal is to determine the subset in $\mathcal{O}$ that contains the $n$-dimensional vector of the means. The problem generalizes most pure exploration bandit problems studied in the literature. We provide the first nearly optimal sample complexity upper and lower bounds for the problem.

cs.LG↗

arXiv2017

The statistical properties of spiral- and scroll-wave turbulence in cardiac tissue

Disorganized electrical activity in the heart leads to sudden cardiac death. To what extent can this electrical turbulence be viewed as classical fluid turbulence,which is an important central problem in modern physics? We investigate,for the first time,via extensive DNSs,the statistical properties of spiral-and scroll-wave turbulence in two- and three-dimensional excitable media by using approaches employed in studies of classical turbulence. We use the Panfilov and the Aliev-Panfilov mathematical models for cardiac tissue. We show that once electrical-wave turbulence has been initiated,there is a forward cascade,in which spirals or scrolls form,interact,and break to yield a turbulent state that is statistically steady and,far away from boundaries,is statistically homogeneous and isotropic. For the transmembrane potential $V$ and the slow recovery variable $g$,which define our models,we define $E_V(k)$ and $E_g(k)$,the electrical-wave analogs of the fluid energy spectrum $E(k)$ in fluid turbulence. We show that $E_V(k)$ and $E_g(k)$ are spread out over several decades in $k$. Thus spiral- and scroll-wave turbulence involves a wide range of spatial scales. $E_V(k)$ and $E_g(k)$ show approximate power laws,in some range of $k$, however,their exponents cannot be determined as accurately as their fluid-turbulence counterparts. The dimensionless ratio $L/λ$ is a convenient control parameter like the Reynolds number for fluid turbulence,where $L$ is the linear size of the domain and $λ$ the wavelength of a plane wave in the medium. By comparing several other statistical properties for spiral- and scroll-wave turbulence with their fluid-turbulence counterparts,we show that,although spiral- and scroll-wave turbulence have some statistical properties like those of fluid turbulence,overall these types of turbulence are special and differ in important ways from fluid turbulence.

physics.bio-ph↗

arXiv2017

Lyapunov dimension of elastic turbulence

Low-Reynolds-number polymer solutions exhibit a chaotic behaviour known as 'elastic turbulence' when the Weissenberg number exceeds a critical value. The two-dimensional Oldroyd-B model is the simplest constitutive model that reproduces this phenomenon. To make a practical estimate of the resolution scale of the dynamics requires an assumption that an attractor of the Oldroyd-B model exists : numerical simulations show that the quantities on which this assumption is based are bounded. We estimate the Lyapunov dimension of this assumed attractor as a function of the Weissenberg number by combining a mathematical analysis of the model with direct numerical simulations.

physics.flu-dyn↗

arXiv2017

Approximation Algorithms for Optimal Decision Trees and Adaptive TSP Problems

We consider the problem of constructing optimal decision trees: given a collection of tests which can disambiguate between a set of $m$ possible diseases, each test having a cost, and the a-priori likelihood of the patient having any particular disease, what is a good adaptive strategy to perform these tests to minimize the expected cost to identify the disease? We settle the approximability of this problem by giving a tight $O(\log m)$-approximation algorithm. We also consider a more substantial generalization, the Adaptive TSP problem. Given an underlying metric space, a random subset $S$ of cities is drawn from a known distribution, but $S$ is initially unknown to us--we get information about whether any city is in $S$ only when we visit the city in question. What is a good adaptive way of visiting all the cities in the random subset $S$ while minimizing the expected distance traveled? For this problem, we give the first poly-logarithmic approximation, and show that this algorithm is best possible unless we can improve the approximation guarantees for the well-known group Steiner tree problem.

cs.DS↗

arXiv2017

Melting of a nonequilibrium vortex crystal in a fluid film with polymers : elastic versus fluid turbulence

We perform a direct numerical simulation (DNS) of the forced, incompressible two-dimensional Navier-Stokes equation coupled with the FENE-P equations for the polymer-conformation tensor. The forcing is such that, without polymers and at low Reynolds numbers $\mbox{Re}$, the film attains a steady state that is a square lattice of vortices and anti-vortices. We find that, as we increase the Weissenberg number $\mbox{Wi}$, a sequence of nonequilibrium phase transitions transforms this lattice, first to spatially distorted, but temporally steady, crystals and then to a sequence of crystals that oscillate in time, periodically, at low $\mbox{Wi}$, and quasiperiodically, for slightly larger $\mbox{Wi}$. Finally, the system becomes disordered and displays spatiotemporal chaos and elastic turbulence. We then obtain the nonequilibrium phase diagram for this system, in the $\mbox{Wi} - Ω$ plane, where $Ω\propto {\mbox{Re}}$, and show that (a) the boundary between the crystalline and turbulent phases has a complicated, fractal-type character and (b) the Okubo-Weiss parameter $Λ$ provides us with a natural measure for characterizing the phases and transitions in this diagram.

physics.flu-dyn↗

arXiv2016

A regularity criterion for solutions of the three-dimensional Cahn-Hilliard-Navier-Stokes equations and associated computations

We consider the 3D Cahn-Hilliard equations coupled to, and driven by, the forced, incompressible 3D Navier-Stokes equations. The combination, known as the Cahn-Hilliard-Navier-Stokes (CHNS) equations, is used in statistical mechanics to model the motion of a binary fluid. The potential development of singularities (blow-up) in the contours of the order parameter $ϕ$ is an open problem. To address this we have proved a theorem that closely mimics the Beale-Kato-Majda theorem for the $3D$ incompressible Euler equations [Beale et al. Commun. Math. Phys., Commun. Math. Phys., ${\rm 94}$, $ 61-66 ({\rm 1984})$]. By taking an $L^{\infty}$ norm of the energy of the full binary system, designated as $E_{\infty}$, we have shown that $\int_{0}^{t}E_{\infty}(τ)\,dτ$ governs the regularity of solutions of the full 3D system. Our direct numerical simulations (DNSs), of the 3D CHNS equations, for (a) a gravity-driven Rayleigh Taylor instability and (b) a constant-energy-injection forcing, with $128^3$ to $512^3$ collocation points and over the duration of our DNSs, confirm that $E_{\infty}$ remains bounded as far as our computations allow.

physics.flu-dyn↗