arXiv subjects

Kannan Ramchandran

Publications and source records attributed to Kannan Ramchandran.

At least 91 records · Page 5Linked to original sources

arXiv2018

Sub-linear Time Support Recovery for Compressed Sensing using Sparse-Graph Codes

We study the support recovery problem for compressed sensing, where the goal is to reconstruct the a high-dimensional $K$-sparse signal $\mathbf{x}\in\mathbb{R}^N$, from low-dimensional linear measurements with and without noise. Our key contribution is a new compressed sensing framework through a new family of carefully designed sparse measurement matrices associated with minimal measurement costs and a low-complexity recovery algorithm. The measurement matrix in our framework is designed based on the well-crafted sparsification through capacity-approaching sparse-graph codes, where the sparse coefficients can be recovered efficiently in a few iterations by performing simple error decoding over the observations. We formally connect this general recovery problem with sparse-graph decoding in packet communication systems, and analyze our framework in terms of the measurement cost, time complexity and recovery performance. In the noiseless setting, our framework can recover any arbitrary $K$-sparse signal in $O(K)$ time using $2K$ measurements asymptotically with high probability. In the noisy setting, when the sparse coefficients take values in a finite and quantized alphabet, our framework can achieve the same goal in time $O(K\log(N/K))$ using $O(K\log(N/K))$ measurements obtained from measurement matrix with elements $\{-1,0,1\}$. When the sparsity $K$ is sub-linear in the signal dimension $K=O(N^δ)$ for some $0<δ<1$, our results are order-optimal in terms of measurement costs and run-time, both of which are sub-linear in the signal dimension $N$. The sub-linear measurement cost and run-time can also be achieved with continuous-valued sparse coefficients, with a slight increment in the logarithmic factors. This offers the desired scalability of our framework that can potentially enable real-time or near-real-time processing for massive datasets featuring sparsity.

cs.IT↗

arXiv2018

Speeding Up Distributed Machine Learning Using Codes

Codes are widely used in many engineering applications to offer robustness against noise. In large-scale systems there are several types of noise that can affect the performance of distributed machine learning algorithms -- straggler nodes, system failures, or communication bottlenecks -- but there has been little interaction cutting across codes, machine learning, and distributed systems. In this work, we provide theoretical insights on how coded solutions can achieve significant gains compared to uncoded ones. We focus on two of the most basic building blocks of distributed learning algorithms: matrix multiplication and data shuffling. For matrix multiplication, we use codes to alleviate the effect of stragglers, and show that if the number of homogeneous workers is $n$, and the runtime of each subtask has an exponential tail, coded computation can speed up distributed matrix multiplication by a factor of $\log n$. For data shuffling, we use codes to reduce communication bottlenecks, exploiting the excess in storage. We show that when a constant fraction $α$ of the data matrix can be cached at each worker, and $n$ is the number of workers, \emph{coded shuffling} reduces the communication cost by a factor of $(α+ \frac{1}{n})γ(n)$ compared to uncoded shuffling, where $γ(n)$ is the ratio of the cost of unicasting $n$ messages to $n$ users to multicasting a common message (of the same size) to $n$ users. For instance, $γ(n) \simeq n$ if multicasting a message to $n$ users is as cheap as unicasting a message to one user. We also provide experiment results, corroborating our theoretical gains of the coded algorithms.

cs.DC↗

arXiv2018

Gradient Diversity: a Key Ingredient for Scalable Distributed Learning

It has been experimentally observed that distributed implementations of mini-batch stochastic gradient descent (SGD) algorithms exhibit speedup saturation and decaying generalization ability beyond a particular batch-size. In this work, we present an analysis hinting that high similarity between concurrently processed gradients may be a cause of this performance degradation. We introduce the notion of gradient diversity that measures the dissimilarity between concurrent gradient updates, and show its key role in the performance of mini-batch SGD. We prove that on problems with high gradient diversity, mini-batch SGD is amenable to better speedups, while maintaining the generalization performance of serial (one sample) SGD. We further establish lower bounds on convergence where mini-batch SGD slows down beyond a particular batch-size, solely due to the lack of gradient diversity. We provide experimental evidence indicating the key role of gradient diversity in distributed learning, and discuss how heuristics like dropout, Langevin dynamics, and quantization can improve it.

cs.LG↗

arXiv2018

Approximate Ranking from Pairwise Comparisons

A common problem in machine learning is to rank a set of n items based on pairwise comparisons. Here ranking refers to partitioning the items into sets of pre-specified sizes according to their scores, which includes identification of the top-k items as the most prominent special case. The score of a given item is defined as the probability that it beats a randomly chosen other item. Finding an exact ranking typically requires a prohibitively large number of comparisons, but in practice, approximate rankings are often adequate. Accordingly, we study the problem of finding approximate rankings from pairwise comparisons. We analyze an active ranking algorithm that counts the number of comparisons won, and decides whether to stop or which pair of items to compare next, based on confidence intervals computed from the data collected in previous steps. We show that this algorithm succeeds in recovering approximate rankings using a number of comparisons that is close to optimal up to logarithmic factors. We also present numerical results, showing that in practice, approximation can drastically reduce the number of comparisons required to estimate a ranking.

cs.LG↗

arXiv2017

A Sequential Approximation Framework for Coded Distributed Optimization

Building on the previous work of Lee et al. and Ferdinand et al. on coded computation, we propose a sequential approximation framework for solving optimization problems in a distributed manner. In a distributed computation system, latency caused by individual processors ("stragglers") usually causes a significant delay in the overall process. The proposed method is powered by a sequential computation scheme, which is designed specifically for systems with stragglers. This scheme has the desirable property that the user is guaranteed to receive useful (approximate) computation results whenever a processor finishes its subtask, even in the presence of uncertain latency. In this paper, we give a coding theorem for sequentially computing matrix-vector multiplications, and the optimality of this coding scheme is also established. As an application of the results, we demonstrate solving optimization problems using a sequential approximation approach, which accelerates the algorithm in a distributed system with stragglers.

cs.IT↗

arXiv2017

The Sample Complexity of Online One-Class Collaborative Filtering

We consider the online one-class collaborative filtering (CF) problem that consists of recommending items to users over time in an online fashion based on positive ratings only. This problem arises when users respond only occasionally to a recommendation with a positive rating, and never with a negative one. We study the impact of the probability of a user responding to a recommendation, p_f, on the sample complexity, i.e., the number of ratings required to make `good' recommendations, and ask whether receiving positive and negative ratings, instead of positive ratings only, improves the sample complexity. Both questions arise in the design of recommender systems. We introduce a simple probabilistic user model, and analyze the performance of an online user-based CF algorithm. We prove that after an initial cold start phase, where recommendations are invested in exploring the user's preferences, this algorithm makes---up to a fraction of the recommendations required for updating the user's preferences---perfect recommendations. The number of ratings required for the cold start phase is nearly proportional to 1/p_f, and that for updating the user's preferences is essentially independent of p_f. As a consequence we find that, receiving positive and negative ratings instead of only positive ones improves the number of ratings required for initial exploration by a factor of 1/p_f, which can be significant.

cs.LG↗

arXiv2017

Fundamental Limits of DNA Storage Systems

Due to its longevity and enormous information density, DNA is an attractive medium for archival storage. In this work, we study the fundamental limits and tradeoffs of DNA-based storage systems under a simple model, motivated by current technological constraints on DNA synthesis and sequencing. Our model captures two key distinctive aspects of DNA storage systems: (1) the data is written onto many short DNA molecules that are stored in an unordered way and (2) the data is read by randomly sampling from this DNA pool. Under this model, we characterize the storage capacity, and show that a simple index-based coding scheme is optimal.

cs.IT↗

arXiv2017

Density Evolution on a Class of Smeared Random Graphs: A Theoretical Framework for Fast MRI

We introduce a new ensemble of random bipartite graphs, which we term the `smearing ensemble', where each left node is connected to some number of consecutive right nodes. Such graphs arise naturally in the recovery of sparse wavelet coefficients when signal acquisition is in the Fourier domain, such as in magnetic resonance imaging (MRI). Graphs from this ensemble exhibit small, structured cycles with high probability, rendering current techniques for determining iterative decoding thresholds inapplicable. In this paper, we develop a theoretical platform to analyze and evaluate the effects of smearing-based structure. Despite the existence of these small cycles, we derive exact density evolution recurrences for iterative decoding on graphs with smear-length two. Further, we give lower bounds on the performance of a much larger class from the smearing ensemble, and provide numerical experiments showing tight agreement between empirical thresholds and those determined by our bounds. Finally, we describe a system architecture to recover sparse wavelet representations in the MRI setting, giving explicit thresholds on the minimum number of Fourier samples needing to be acquired for the $1$-stage Haar wavelet setting. In particular, we show that $K$-sparse $1$-stage Haar wavelet coefficients of an $n$-dimensional signal can be recovered using $2.63K$ Fourier domain samples asymptotically using $\mathcal{O}(K\log{K})$ operations.

cs.IT↗

arXiv2017

PhaseCode: Fast and Efficient Compressive Phase Retrieval based on Sparse-Graph-Codes

We consider the problem of recovering a $K$-sparse complex signal $x$ from $m$ intensity measurements. We propose the PhaseCode algorithm, and show that in the noiseless case, PhaseCode can recover an arbitrarily-close-to-one fraction of the $K$ non-zero signal components using only slightly more than $4K$ measurements when the support of the signal is uniformly random, with order-optimal time and memory complexity of $Θ(K)$. It is known that the fundamental limit for the number of measurements in compressive phase retrieval problem is $4K - o(K)$ to recover the signal exactly and with no assumptions on its support distribution. This shows that under mild relaxation of the conditions, our algorithm is the first constructive \emph{capacity-approaching} compressive phase retrieval algorithm: in fact, our algorithm is also order-optimal in complexity and memory. Next, motivated by some important practical classes of optical systems, we consider a Fourier-friendly constrained measurement setting, and show that its performance matches that of the unconstrained setting. In the Fourier-friendly setting that we consider, the measurement matrix is constrained to be a cascade of Fourier matrices and diagonal matrices. We further demonstrate how PhaseCode can be robustified to noise. Throughout, we provide extensive simulation results that validate the practical power of our proposed algorithms for the sparse unconstrained and Fourier-friendly measurement settings, for noiseless and noisy scenarios. A key contribution of our work is the novel use of coding-theoretic tools like density evolution methods for the design and analysis of fast and efficient algorithms for compressive phase-retrieval problems.

cs.IT↗

arXiv2016

Active Ranking from Pairwise Comparisons and when Parametric Assumptions Don't Help

We consider sequential or active ranking of a set of n items based on noisy pairwise comparisons. Items are ranked according to the probability that a given item beats a randomly chosen item, and ranking refers to partitioning the items into sets of pre-specified sizes according to their scores. This notion of ranking includes as special cases the identification of the top-k items and the total ordering of the items. We first analyze a sequential ranking algorithm that counts the number of comparisons won, and uses these counts to decide whether to stop, or to compare another pair of items, chosen based on confidence intervals specified by the data collected up to that point. We prove that this algorithm succeeds in recovering the ranking using a number of comparisons that is optimal up to logarithmic factors. This guarantee does not require any structural properties of the underlying pairwise probability matrix, unlike a significant body of past work on pairwise ranking based on parametric models such as the Thurstone or Bradley-Terry-Luce models. It has been a long-standing open question as to whether or not imposing these parametric assumptions allows for improved ranking algorithms. For stochastic comparison models, in which the pairwise probabilities are bounded away from zero, our second contribution is to resolve this issue by proving a lower bound for parametric models. This shows, perhaps surprisingly, that these popular parametric modeling choices offer at most logarithmic gains for stochastic comparisons.

cs.LG↗

arXiv2016

Hiding the Rumor Source

Anonymous social media platforms like Secret, Yik Yak, and Whisper have emerged as important tools for sharing ideas without the fear of judgment. Such anonymous platforms are also important in nations under authoritarian rule, where freedom of expression and the personal safety of message authors may depend on anonymity. Whether for fear of judgment or retribution, it is sometimes crucial to hide the identities of users who post sensitive messages. In this paper, we consider a global adversary who wishes to identify the author of a message; it observes either a snapshot of the spread of a message at a certain time, sampled timestamp metadata, or both. Recent advances in rumor source detection show that existing messaging protocols are vulnerable against such an adversary. We introduce a novel messaging protocol, which we call adaptive diffusion, and show that under the snapshot adversarial model, adaptive diffusion spreads content fast and achieves perfect obfuscation of the source when the underlying contact network is an infinite regular tree. That is, all users with the message are nearly equally likely to have been the origin of the message. When the contact network is an irregular tree, we characterize the probability of maximum likelihood detection by proving a concentration result over Galton-Watson trees. Experiments on a sampled Facebook network demonstrate that adaptive diffusion effectively hides the location of the source even when the graph is finite, irregular and has cycles.

cs.CR↗

arXiv2016

Fast and Robust Compressive Phase Retrieval with Sparse-Graph Codes

In this paper, we tackle the compressive phase retrieval problem in the presence of noise. The noisy compressive phase retrieval problem is to recover a $K$-sparse complex signal $s \in \mathbb{C}^n$, from a set of $m$ noisy quadratic measurements: $ y_i=| a_i^H s |^2+w_i$, where $a_i^H\in\mathbb{C}^n$ is the $i$th row of the measurement matrix $A\in\mathbb{C}^{m\times n}$, and $w_i$ is the additive noise to the $i$th measurement. We consider the regime where $K=βn^δ$, with constants $β>0$ and $δ\in(0,1)$. We use the architecture of PhaseCode algorithm, and robustify it using two schemes: the almost-linear scheme and the sublinear scheme. We prove that with high probability, the almost-linear scheme recovers $s$ with sample complexity $Θ(K \log(n))$ and computational complexity $Θ(n \log(n))$, and the sublinear scheme recovers $s$ with sample complexity $Θ(K\log^3(n))$ and computational complexity $Θ(K\log^3(n))$. To the best of our knowledge, this is the first scheme that achieves sublinear computational complexity for compressive phase retrieval problem. Finally, we provide simulation results that support our theoretical contributions.

cs.IT↗

arXiv2016

CYCLADES: Conflict-free Asynchronous Machine Learning

We present CYCLADES, a general framework for parallelizing stochastic optimization algorithms in a shared memory setting. CYCLADES is asynchronous during shared model updates, and requires no memory locking mechanisms, similar to HOGWILD!-type algorithms. Unlike HOGWILD!, CYCLADES introduces no conflicts during the parallel execution, and offers a black-box analysis for provable speedups across a large family of algorithms. Due to its inherent conflict-free nature and cache locality, our multi-core implementation of CYCLADES consistently outperforms HOGWILD!-type algorithms on sufficiently sparse datasets, leading to up to 40% speedup gains compared to the HOGWILD! implementation of SGD, and up to 5x gains over asynchronous implementations of variance reduction algorithms.

stat.ML↗

arXiv2016

Perturbed Iterate Analysis for Asynchronous Stochastic Optimization

We introduce and analyze stochastic optimization methods where the input to each gradient update is perturbed by bounded noise. We show that this framework forms the basis of a unified approach to analyze asynchronous implementations of stochastic optimization algorithms.In this framework, asynchronous stochastic optimization algorithms can be thought of as serial methods operating on noisy inputs. Using our perturbed iterate framework, we provide new analyses of the Hogwild! algorithm and asynchronous stochastic coordinate descent, that are simpler than earlier analyses, remove many assumptions of previous models, and in some cases yield improved upper bounds on the convergence rates. We proceed to apply our framework to develop and analyze KroMagnon: a novel, parallel, sparse stochastic variance-reduced gradient (SVRG) algorithm. We demonstrate experimentally on a 16-core machine that the sparse and parallel version of SVRG is in some cases more than four orders of magnitude faster than the standard SVRG algorithm.

stat.ML↗

arXiv2015

Fast and Efficient Sparse 2D Discrete Fourier Transform using Sparse-Graph Codes

We present a novel algorithm, named the 2D-FFAST, to compute a sparse 2D-Discrete Fourier Transform (2D-DFT) featuring both low sample complexity and low computational complexity. The proposed algorithm is based on mixed concepts from signal processing (sub-sampling and aliasing), coding theory (sparse-graph codes) and number theory (Chinese-remainder-theorem) and generalizes the 1D-FFAST 2 algorithm recently proposed by Pawar and Ramchandran [1] to the 2D setting. Concretely, our proposed 2D-FFAST algorithm computes a k-sparse 2D-DFT, with a uniformly random support, of size N = Nx x Ny using O(k) noiseless spatial-domain measurements in O(k log k) computational time. Our results are attractive when the sparsity is sub-linear with respect to the signal dimension, that is, when k -> infinity and k/N -> 0. For the case when the spatial-domain measurements are corrupted by additive noise, our 2D-FFAST framework extends to a noise-robust version in sub-linear time of O(k log4 N ) using O(k log3 N ) measurements. Simulation results, on synthetic images as well as real-world magnetic resonance images, are provided in Section VII and demonstrate the empirical performance of the proposed 2D-FFAST algorithm.

cs.IT↗

arXiv2015

Low-Complexity Interactive Algorithms for Synchronization from Deletions, Insertions, and Substitutions

Consider two remote nodes having binary sequences $X$ and $Y$, respectively. $Y$ is an edited version of ${X}$, where the editing involves random deletions, insertions, and substitutions, possibly in bursts. The goal is for the node with $Y$ to reconstruct $X$ with minimal exchange of information over a noiseless link. The communication is measured in terms of both the total number of bits exchanged and the number of interactive rounds of communication. This paper focuses on the setting where the number of edits is $o(\tfrac{n}{\log n})$, where $n$ is the length of $X$. We first consider the case where the edits are a mixture of insertions and deletions (indels), and propose an interactive synchronization algorithm with near-optimal communication rate and average computational complexity of $O(n)$ arithmetic operations. The algorithm uses interaction to efficiently split the source sequence into substrings containing exactly one deletion or insertion. Each of these substrings is then synchronized using an optimal one-way synchronization code based on the single-deletion correcting channel codes of Varshamov and Tenengolts (VT codes). We then build on this synchronization algorithm in three different ways. First, it is modified to work with a single round of interaction. The reduction in the number of rounds comes at the expense of higher communication, which is quantified. Next, we present an extension to the practically important case where the insertions and deletions may occur in (potentially large) bursts. Finally, we show how to synchronize the sources to within a target Hamming distance. This feature can be used to differentiate between substitution and indel edits. In addition to theoretical performance bounds, we provide several validating simulation results for the proposed algorithms.

cs.IT↗

arXiv2015

Optimal Systematic Distributed Storage Codes with Fast Encoding

Erasure codes are being increasingly used in distributed-storage systems in place of data-replication, since they provide the same level of reliability with much lower storage overhead. We consider the problem of constructing explicit erasure codes for distributed storage with the following desirable properties motivated by practice: (i) Maximum-Distance-Separable (MDS): to provide maximal reliability at minimum storage overhead, (ii) Optimal repair-bandwidth: to minimize the amount of data needed to be transferred to repair a failed node from remaining ones, (iii) Flexibility in repair: to allow maximal flexibility in selecting subset of nodes to use for repair, which includes not requiring that all surviving nodes be used for repair, (iv) Systematic Form: to ensure that the original data exists in uncoded form, and (v) Fast encoding: to minimize the cost of generating encoded data (enabled by a sparse generator matrix). This paper presents the first explicit code construction which theoretically guarantees all the five desired properties simultaneously. Our construction builds on a powerful class of codes called Product-Matrix (PM) codes. PM codes satisfy properties (i)-(iii), and either (iv) or (v), but not both simultaneously. Indeed, native PM codes have inherent structure that leads to sparsity, but this structure is destroyed when the codes are made systematic. We first present an analytical framework for understanding the interaction between the design of PM codes and the systematic property. Using this framework, we provide an explicit code construction that simultaneously achieves all the above desired properties. We also present general ways of transforming existing storage and repair optimal codes to enable fast encoding through sparsity. In practice, such sparse codes result in encoding speedup by a factor of about 4 for typical parameters.

cs.IT↗

arXiv2015

SPRIGHT: A Fast and Robust Framework for Sparse Walsh-Hadamard Transform

We consider the problem of computing the Walsh-Hadamard Transform (WHT) of some $N$-length input vector in the presence of noise, where the $N$-point Walsh spectrum is $K$-sparse with $K = {O}(N^δ)$ scaling sub-linearly in the input dimension $N$ for some $0<δ<1$. Over the past decade, there has been a resurgence in research related to the computation of Discrete Fourier Transform (DFT) for some length-$N$ input signal that has a $K$-sparse Fourier spectrum. In particular, through a sparse-graph code design, our earlier work on the Fast Fourier Aliasing-based Sparse Transform (FFAST) algorithm computes the $K$-sparse DFT in time ${O}(K\log K)$ by taking ${O}(K)$ noiseless samples. Inspired by the coding-theoretic design framework, Scheibler et al. proposed the Sparse Fast Hadamard Transform (SparseFHT) algorithm that elegantly computes the $K$-sparse WHT in the absence of noise using ${O}(K\log N)$ samples in time ${O}(K\log^2 N)$. However, the SparseFHT algorithm explicitly exploits the noiseless nature of the problem, and is not equipped to deal with scenarios where the observations are corrupted by noise. Therefore, a question of critical interest is whether this coding-theoretic framework can be made robust to noise. Further, if the answer is yes, what is the extra price that needs to be paid for being robust to noise? In this paper, we show, quite interestingly, that there is {\it no extra price} that needs to be paid for being robust to noise other than a constant factor. In other words, we can maintain the same sample complexity ${O}(K\log N)$ and the computational complexity ${O}(K\log^2 N)$ as those of the noiseless case, using our SParse Robust Iterative Graph-based Hadamard Transform (SPRIGHT) algorithm.

cs.IT↗