arXiv subjects

Atri Rudra

Publications and source records attributed to Atri Rudra.

At least 55 records · Page 3Linked to original sources

arXiv2018

Hypertree Decompositions Revisited for PGMs

We revisit the classical problem of exact inference on probabilistic graphical models (PGMs). Our algorithm is based on recent worst-case optimal database join algorithms, which can be asymptotically faster than traditional data processing methods. We present the first empirical evaluation of these new algorithms via JoinInfer, a new exact inference engine. We empirically explore the properties of the data for which our engine can be expected to outperform traditional inference engines refining current theoretical notions. Further, JoinInfer outperforms existing state-of-the-art inference engines (ACE, IJGP and libDAI) on some standard benchmark datasets by up to a factor of 630x. Finally, we propose a promising data-driven heuristic that extends JoinInfer to automatically tailor its parameters and/or switch to the traditional inference algorithms.

cs.AI↗

arXiv2017

A Two Pronged Progress in Structured Dense Matrix Multiplication

Matrix-vector multiplication is one of the most fundamental computing primitives. Given a matrix $A\in\mathbb{F}^{N\times N}$ and a vector $b$, it is known that in the worst case $Θ(N^2)$ operations over $\mathbb{F}$ are needed to compute $Ab$. A broad question is to identify classes of structured dense matrices that can be represented with $O(N)$ parameters, and for which matrix-vector multiplication can be performed sub-quadratically. One such class of structured matrices is the orthogonal polynomial transforms, whose rows correspond to a family of orthogonal polynomials. Other well known classes include the Toeplitz, Hankel, Vandermonde, Cauchy matrices and their extensions that are all special cases of a ldisplacement rank property. In this paper, we make progress on two fronts: 1. We introduce the notion of recurrence width of matrices. For matrices with constant recurrence width, we design algorithms to compute $Ab$ and $A^Tb$ with a near-linear number of operations. This notion of width is finer than all the above classes of structured matrices and thus we can compute multiplication for all of them using the same core algorithm. 2. We additionally adapt this algorithm to an algorithm for a much more general class of matrices with displacement structure: those with low displacement rank with respect to quasiseparable matrices. This class includes Toeplitz-plus-Hankel-like matrices, Discrete Cosine/Sine Transforms, and more, and captures all previously known matrices with displacement structure that we are aware of under a unified parametrization and algorithm. Our work unifies, generalizes, and simplifies existing state-of-the-art results in structured matrix-vector multiplication. Finally, we show how applications in areas such as multipoint evaluations of multivariate polynomials can be reduced to problems involving low recurrence width matrices.

cs.DS↗

arXiv2017

Average-radius list-recovery of random linear codes: it really ties the room together

We analyze the list-decodability, and related notions, of random linear codes. This has been studied extensively before: there are many different parameter regimes and many different variants. Previous works have used complementary styles of arguments---which each work in their own parameter regimes but not in others---and moreover have left some gaps in our understanding of the list-decodability of random linear codes. In particular, none of these arguments work well for list-recovery, a generalization of list-decoding that has been useful in a variety of settings. In this work, we present a new approach, which works across parameter regimes and further generalizes to list-recovery. Our main theorem can establish better list-decoding and list-recovery results for low-rate random linear codes over large fields; list-recovery of high-rate random linear codes; and it can recover the rate bounds of Guruswami, Hastad, and Kopparty for constant-rate random linear codes (although with large list sizes).

cs.IT↗

arXiv2017

Juggling Functions Inside a Database

We define and study the Functional Aggregate Query (FAQ) problem, which captures common computational tasks across a very wide range of domains including relational databases, logic, matrix and tensor computation, probabilistic graphical models, constraint satisfaction, and signal processing. Simply put, an FAQ is a declarative way of defining a new function from a database of input functions. We present "InsideOut", a dynamic programming algorithm, to evaluate an FAQ. The algorithm rewrites the input query into a set of easier-to-compute FAQ sub-queries. Each sub-query is then evaluated using a worst-case optimal relational join algorithm. The topic of designing algorithms to optimally evaluate the classic multiway join problem has seen exciting developments in the past few years. Our framework tightly connects these new ideas in database theory with a vast number of application areas in a coherent manner, showing potentially that a good database engine can be a general-purpose constraint solver, relational data store, graphical model inference engine, and matrix/tensor computation processor all at once. The InsideOut algorithm is very simple, as shall be described in this paper. Yet, in spite of solving an extremely general problem, its runtime either is as good as or improves upon the best known algorithm for the applications that FAQ specializes to. These corollaries include computational tasks in graphical model inference, matrix/tensor operations, relational joins, and logic. Better yet, InsideOut can be used within any database engine, because it is basically a principled way of rewriting queries. Indeed, it is already part of the LogicBlox database engine, helping efficiently answer traditional database queries, graphical model inference queries, and train a large class of machine learning models inside the database itself.

cs.DB↗

arXiv2017

Implementation of Tetris as a Model Counter

Solving SharpSAT problems is an important area of work. In this paper, we discuss implementing Tetris, an algorithm originally designed for handling natural joins, as an exact model counter for the SharpSAT problem. Tetris uses a simple geometric framework, yet manages to achieve the fractional hypertree-width bound. Its design allows it to handle complex problems involving extremely large numbers of clauses on which other state-of-the-art model counters do not perform well, yet still performs strongly on standard SAT benchmarks. We have achieved the following objectives. First, we have found a natural set of model counting benchmarks on which Tetris outperforms other model counters. Second, we have constructed a data structure capable of efficiently handling and caching all of the data Tetris needs to work on over the course of the algorithm. Third, we have modified Tetris in order to move from a theoretical, asymptotic-time-focused environment to one that performs well in practice. In particular, we have managed to produce results keeping us within a single order of magnitude as compared to other solvers on most benchmarks, and outperform those solvers by multiple orders of magnitude on others.

cs.DS↗

arXiv2016

Joins via Geometric Resolutions: Worst-case and Beyond

We present a simple geometric framework for the relational join. Using this framework, we design an algorithm that achieves the fractional hypertree-width bound, which generalizes classical and recent worst-case algorithmic results on computing joins. In addition, we use our framework and the same algorithm to show a series of what are colloquially known as beyond worst-case results. The framework allows us to prove results for data stored in Btrees, multidimensional data structures, and even multiple indices per table. A key idea in our framework is formalizing the inference one does with an index as a type of geometric resolution; transforming the algorithmic problem of computing joins to a geometric problem. Our notion of geometric resolution can be viewed as a geometric analog of logical resolution. In addition to the geometry and logic connections, our algorithm can also be thought of as backtracking search with memoization.

cs.DB↗

arXiv2016

Tight Network Topology Dependent Bounds on Rounds of Communication

We prove tight network topology dependent bounds on the round complexity of computing well studied $k$-party functions such as set disjointness and element distinctness. Unlike the usual case in the CONGEST model in distributed computing, we fix the function and then vary the underlying network topology. This complements the recent such results on total communication that have received some attention. We also present some applications to distributed graph computation problems. Our main contribution is a proof technique that allows us to reduce the problem on a general graph topology to a relevant two-party communication complexity problem. However, unlike many previous works that also used the same high level strategy, we do not reason about a two-party communication problem that is induced by a cut in the graph. To `stitch' back the various lower bounds from the two party communication problems, we use the notion of timed graph that has seen prior use in network coding. Our reductions use some tools from Steiner tree packing and multi-commodity flow problems that have a delay constraint.

cs.CC↗

arXiv2015

The Range of Topological Effects on Communication

We continue the study of communication cost of computing functions when inputs are distributed among $k$ processors, each of which is located at one vertex of a network/graph called a terminal. Every other node of the network also has a processor, with no input. The communication is point-to-point and the cost is the total number of bits exchanged by the protocol, in the worst case, on all edges. Chattopadhyay, Radhakrishnan and Rudra (FOCS'14) recently initiated a study of the effect of topology of the network on the total communication cost using tools from $L_1$ embeddings. Their techniques provided tight bounds for simple functions like Element-Distinctness (ED), which depend on the 1-median of the graph. This work addresses two other kinds of natural functions. We show that for a large class of natural functions like Set-Disjointness the communication cost is essentially $n$ times the cost of the optimal Steiner tree connecting the terminals. Further, we show for natural composed functions like $\text{ED} \circ \text{XOR}$ and $\text{XOR} \circ \text{ED}$, the naive protocols suggested by their definition is optimal for general networks. Interestingly, the bounds for these functions depend on more involved topological parameters that are a combination of Steiner tree and 1-median costs. To obtain our results, we use some new tools in addition to ones used in Chattopadhyay et. al. These include (i) viewing the communication constraints via a linear program; (ii) using tools from the theory of tree embeddings to prove topology sensitive direct sum results that handle the case of composed functions and (iii) representing the communication constraints of certain problems as a family of collection of multiway cuts, where each multiway cut simulates the hardness of computing the function on the star topology.

cs.CC↗

arXiv2015

Join Processing for Graph Patterns: An Old Dog with New Tricks

Join optimization has been dominated by Selinger-style, pairwise optimizers for decades. But, Selinger-style algorithms are asymptotically suboptimal for applications in graphic analytics. This suboptimality is one of the reasons that many have advocated supplementing relational engines with specialized graph processing engines. Recently, new join algorithms have been discovered that achieve optimal worst-case run times for any join or even so-called beyond worst-case (or instance optimal) run time guarantees for specialized classes of joins. These new algorithms match or improve on those used in specialized graph-processing systems. This paper asks can these new join algorithms allow relational engines to close the performance gap with graph engines? We examine this question for graph-pattern queries or join queries. We find that classical relational databases like Postgres and MonetDB or newer graph databases/stores like Virtuoso and Neo4j may be orders of magnitude slower than these new approaches compared to a fully featured RDBMS, LogicBlox, using these new ideas. Our results demonstrate that an RDBMS with such new algorithms can perform as well as specialized engines like GraphLab -- while retaining a high-level interface. We hope this adds to the ongoing debate of the role of graph accelerators, new graph systems, and relational systems in modern workloads.

cs.DB↗

arXiv2014

It'll probably work out: improved list-decoding through random operations

In this work, we introduce a framework to study the effect of random operations on the combinatorial list-decodability of a code. The operations we consider correspond to row and column operations on the matrix obtained from the code by stacking the codewords together as columns. This captures many natural transformations on codes, such as puncturing, folding, and taking subcodes; we show that many such operations can improve the list-decoding properties of a code. There are two main points to this. First, our goal is to advance our (combinatorial) understanding of list-decodability, by understanding what structure (or lack thereof) is necessary to obtain it. Second, we use our more general results to obtain a few interesting corollaries for list decoding: (1) We show the existence of binary codes that are combinatorially list-decodable from $1/2-ε$ fraction of errors with optimal rate $Ω(ε^2)$ that can be encoded in linear time. (2) We show that any code with $Ω(1)$ relative distance, when randomly folded, is combinatorially list-decodable $1-ε$ fraction of errors with high probability. This formalizes the intuition for why the folding operation has been successful in obtaining codes with optimal list decoding parameters; previously, all arguments used algebraic methods and worked only with specific codes. (3) We show that any code which is list-decodable with suboptimal list sizes has many subcodes which have near-optimal list sizes, while retaining the error correcting capabilities of the original code. This generalizes recent results where subspace evasive sets have been used to reduce list sizes of codes that achieve list decoding capacity.

cs.IT↗

arXiv2014

Sparse Approximation, List Decoding, and Uncertainty Principles

We consider list versions of sparse approximation problems, where unlike the existing results in sparse approximation that consider situations with unique solutions, we are interested in multiple solutions. We introduce these problems and present the first combinatorial results on the output list size. These generalize and enhance some of the existing results on threshold phenomenon and uncertainty principles in sparse approximations. Our definitions and results are inspired by similar results in list decoding. We also present lower bound examples that bolster our results and show they are of the appropriate size.

cs.IT↗

arXiv2014

Beyond Worst-Case Analysis for Joins with Minesweeper

We describe a new algorithm, Minesweeper, that is able to satisfy stronger runtime guarantees than previous join algorithms (colloquially, `beyond worst-case guarantees') for data in indexed search trees. Our first contribution is developing a framework to measure this stronger notion of complexity, which we call {\it certificate complexity}, that extends notions of Barbay et al. and Demaine et al.; a certificate is a set of propositional formulae that certifies that the output is correct. This notion captures a natural class of join algorithms. In addition, the certificate allows us to define a strictly stronger notion of runtime complexity than traditional worst-case guarantees. Our second contribution is to develop a dichotomy theorem for the certificate-based notion of complexity. Roughly, we show that Minesweeper evaluates $β$-acyclic queries in time linear in the certificate plus the output size, while for any $β$-cyclic query there is some instance that takes superlinear time in the certificate (and for which the output is no larger than the certificate size). We also extend our certificate-complexity analysis to queries with bounded treewidth and the triangle query.

cs.DB↗

arXiv2013

Skew Strikes Back: New Developments in the Theory of Join Algorithms

Evaluating the relational join is one of the central algorithmic and most well-studied problems in database systems. A staggering number of variants have been considered including Block-Nested loop join, Hash-Join, Grace, Sort-merge for discussions of more modern issues). Commercial database engines use finely tuned join heuristics that take into account a wide variety of factors including the selectivity of various predicates, memory, IO, etc. In spite of this study of join queries, the textbook description of join processing is suboptimal. This survey describes recent results on join algorithms that have provable worst-case optimality runtime guarantees. We survey recent work and provide a simpler and unified description of these algorithms that we hope is useful for theory-minded readers, algorithm designers, and systems implementors.

cs.DB↗

arXiv2013

Every list-decodable code for high noise has abundant near-optimal rate puncturings

We show that any q-ary code with sufficiently good distance can be randomly punctured to obtain, with high probability, a code that is list decodable up to radius $1 - 1/q - ε$ with near-optimal rate and list sizes. Our results imply that "most" Reed-Solomon codes are list decodable beyond the Johnson bound, settling the long-standing open question of whether any Reed Solomon codes meet this criterion. More precisely, we show that a Reed-Solomon code with random evaluation points is, with high probability, list decodable up to radius $1 - ε$ with list sizes $O(1/ε)$ and rate $Ω(ε)$. As a second corollary of our argument, we obtain improved bounds on the list decodability of random linear codes over large fields. Our approach exploits techniques from high dimensional probability. Previous work used similar tools to obtain bounds on the list decodability of random linear codes, but the bounds did not scale with the size of the alphabet. In this paper, we use a chaining argument to deal with large alphabet sizes.

cs.IT↗

arXiv2013

Accurate Decoding of Pooled Sequenced Data Using Compressed Sensing

In order to overcome the limitations imposed by DNA barcoding when multiplexing a large number of samples in the current generation of high-throughput sequencing instruments, we have recently proposed a new protocol that leverages advances in combinatorial pooling design (group testing) doi:10.1371/journal.pcbi.1003010. We have also demonstrated how this new protocol would enable de novo selective sequencing and assembly of large, highly-repetitive genomes. Here we address the problem of decoding pooled sequenced data obtained from such a protocol. Our algorithm employs a synergistic combination of ideas from compressed sensing and the decoding of error-correcting codes. Experimental results on synthetic data for the rice genome and real data for the barley genome show that our novel decoding algorithm enables significantly higher quality assemblies than the previous approach.

q-bio.QM↗

arXiv2013

L2/L2-foreach sparse recovery with low risk

In this paper, we consider the "foreach" sparse recovery problem with failure probability $p$. The goal of which is to design a distribution over $m \times N$ matrices $Φ$ and a decoding algorithm $\algo$ such that for every $\vx\in\R^N$, we have the following error guarantee with probability at least $1-p$ \[\|\vx-\algo(Φ\vx)\|_2\le C\|\vx-\vx_k\|_2,\] where $C$ is a constant (ideally arbitrarily close to 1) and $\vx_k$ is the best $k$-sparse approximation of $\vx$. Much of the sparse recovery or compressive sensing literature has focused on the case of either $p = 0$ or $p = Ω(1)$. We initiate the study of this problem for the entire range of failure probability. Our two main results are as follows: \begin{enumerate} \item We prove a lower bound on $m$, the number measurements, of $Ω(k\log(n/k)+\log(1/p))$ for $2^{-Θ(N)}\le p <1$. Cohen, Dahmen, and DeVore \cite{CDD2007:NearOptimall2l2} prove that this bound is tight. \item We prove nearly matching upper bounds for \textit{sub-linear} time decoding. Previous such results addressed only $p = Ω(1)$. \end{enumerate} Our results and techniques lead to the following corollaries: (i) the first ever sub-linear time decoding $\lolo$ "forall" sparse recovery system that requires a $\log^γ{N}$ extra factor (for some $γ<1$) over the optimal $O(k\log(N/k))$ number of measurements, and (ii) extensions of Gilbert et al. \cite{GHRSW12:SimpleSignals} results for information-theoretically bounded adversaries.

cs.DS↗

arXiv2012

Almost Universal Hash Families are also Storage Enforcing

We show that every almost universal hash function also has the storage enforcement property. Almost universal hash functions have found numerous applications and we show that this new storage enforcement property allows the application of almost universal hash functions in a wide range of remote verification tasks: (i) Proof of Secure Erasure (where we want to remotely erase and securely update the code of a compromised machine with memory-bounded adversary), (ii) Proof of Ownership (where a storage server wants to check if a client has the data it claims to have before giving access to deduplicated data) and (iii) Data possession (where the client wants to verify whether the remote storage server is storing its data). Specifically, storage enforcement guarantee in the classical data possession problem removes any practical incentive for the storage server to cheat the client by saving on storage space. The proof of our result relies on a natural combination of Kolmogorov Complexity and List Decoding. To the best of our knowledge this is the first work that combines these two techniques. We believe the newly introduced storage enforcement property of almost universal hash functions will open promising avenues of exciting research under memory-bounded (bounded storage) adversary model.

cs.IT↗

arXiv2012

Analyzing Nonblocking Switching Networks using Linear Programming (Duality)

The main task in analyzing a switching network design (including circuit-, multirate-, and photonic-switching) is to determine the minimum number of some switching components so that the design is non-blocking in some sense (e.g., strict- or wide-sense). We show that, in many cases, this task can be accomplished with a simple two-step strategy: (1) formulate a linear program whose optimum value is a bound for the minimum number we are seeking, and (2) specify a solution to the dual program, whose objective value by weak duality immediately yields a sufficient condition for the design to be non-blocking. We illustrate this technique through a variety of examples, ranging from circuit to multirate to photonic switching, from unicast to $f$-cast and multicast, and from strict- to wide-sense non-blocking. The switching architectures in the examples are of Clos-type and Banyan-type, which are the two most popular architectural choices for designing non-blocking switching networks. To prove the result in the multirate Clos network case, we formulate a new problem called {\sc dynamic weighted edge coloring} which generalizes the {\sc dynamic bin packing} problem. We then design an algorithm with competitive ratio 5.6355 for the problem. The algorithm is analyzed using the linear programming technique. A new upper-bound for multirate wide-sense non-blocking Clos networks follow, improving upon a decade-old bound on the same problem.

cs.DM↗