arXiv subjects

Ashish Goel

Publications and source records attributed to Ashish Goel.

At least 73 records · Page 4Linked to original sources

arXiv2014

Crowdsourcing for Participatory Democracies: Efficient Elicitation of Social Choice Functions

We present theoretical and empirical results demonstrating the usefulness of voting rules for participatory democracies. We first give algorithms which efficiently elicit ε-approximations to two prominent voting rules: the Borda rule and the Condorcet winner. This result circumvents previous prohibitive lower bounds and is surprisingly strong: even if the number of ideas is as large as the number of participants, each participant will only have to make a logarithmic number of comparisons, an exponential improvement over the linear number of comparisons previously needed. We demonstrate the approach in an experiment in Finland's recent off-road traffic law reform, observing that the total number of comparisons needed to achieve a fixed εapproximation is linear in the number of ideas and that the constant is not large. Finally, we note a few other experimental observations which support the use of voting rules for aggregation. First, we observe that rating, one of the common alternatives to ranking, manifested effects of bias in our data. Second, we show that very few of the topics lacked a Condorcet winner, one of the prominent negative results in voting. Finally, we show data hinting at a potential future direction: the use of partial rankings as opposed to pairwise comparisons to further decrease the elicitation time.

cs.MA↗

arXiv2014

Personalized PageRank to a Target Node

Personalalized PageRank uses random walks to determine the importance or authority of nodes in a graph from the point of view of a given source node. Much past work has considered how to compute personalized PageRank from a given source node to other nodes. In this work we consider the problem of computing personalized PageRanks to a given target node from all source nodes. This problem can be interpreted as finding who supports the target or who is interested in the target. We present an efficient algorithm for computing personalized PageRank to a given target up to any given accuracy. We give a simple analysis of our algorithm's running time in both the average case and the parameterized worst-case. We show that for any graph with $n$ nodes and $m$ edges, if the target node is randomly chosen and the teleport probability $α$ is given, the algorithm will compute a result with $ε$ error in time $O\left(\frac{1}{αε} \left(\frac{m}{n} + \log(n)\right)\right)$. This is much faster than the previously proposed method of computing personalized PageRank separately from every source node, and it is comparable to the cost of computing personalized PageRank from a single source. We present results from experiments on the Twitter graph which show that the constant factors in our running time analysis are small and our algorithm is efficient in practice.

cs.DS↗

arXiv2013

Dimension Independent Similarity Computation

We present a suite of algorithms for Dimension Independent Similarity Computation (DISCO) to compute all pairwise similarities between very high dimensional sparse vectors. All of our results are provably independent of dimension, meaning apart from the initial cost of trivially reading in the data, all subsequent operations are independent of the dimension, thus the dimension can be very large. We study Cosine, Dice, Overlap, and the Jaccard similarity measures. For Jaccard similiarity we include an improved version of MinHash. Our results are geared toward the MapReduce framework. We empirically validate our theorems at large scale using data from the social networking site Twitter. At time of writing, our algorithms are live in production at twitter.com.

cs.DS↗

arXiv2012

Complexity Measures for Map-Reduce, and Comparison to Parallel Computing

The programming paradigm Map-Reduce and its main open-source implementation, Hadoop, have had an enormous impact on large scale data processing. Our goal in this expository writeup is two-fold: first, we want to present some complexity measures that allow us to talk about Map-Reduce algorithms formally, and second, we want to point out why this model is actually different from other models of parallel programming, most notably the PRAM (Parallel Random Access Memory) model. We are looking for complexity measures that are detailed enough to make fine-grained distinction between different algorithms, but which also abstract away many of the implementation details.

cs.DC↗

arXiv2012

Efficient Distributed Locality Sensitive Hashing

Distributed frameworks are gaining increasingly widespread use in applications that process large amounts of data. One important example application is large scale similarity search, for which Locality Sensitive Hashing (LSH) has emerged as the method of choice, specially when the data is high-dimensional. At its core, LSH is based on hashing the data points to a number of buckets such that similar points are more likely to map to the same buckets. To guarantee high search quality, the LSH scheme needs a rather large number of hash tables. This entails a large space requirement, and in the distributed setting, with each query requiring a network call per hash bucket look up, this also entails a big network load. The Entropy LSH scheme proposed by Panigrahy significantly reduces the number of required hash tables by looking up a number of query offsets in addition to the query itself. While this improves the LSH space requirement, it does not help with (and in fact worsens) the search network efficiency, as now each query offset requires a network call. In this paper, focusing on the Euclidian space under $l_2$ norm and building up on Entropy LSH, we propose the distributed Layered LSH scheme, and prove that it exponentially decreases the network cost, while maintaining a good load balance between different machines. Our experiments also verify that our scheme results in a significant network traffic reduction that brings about large runtime improvement in real world applications.

cs.DC↗

arXiv2012

Triadic Consensus: A Randomized Algorithm for Voting in a Crowd

Typical voting rules do not work well in settings with many candidates. If there are just several hundred candidates, then even a simple task such as choosing a top candidate becomes impractical. Motivated by the hope of developing group consensus mechanisms over the internet, where the numbers of candidates could easily number in the thousands, we study an urn-based voting rule where each participant acts as a voter and a candidate. We prove that when participants lie in a one-dimensional space, this voting protocol finds a $(1-ε/sqrt{n})$ approximation of the Condorcet winner with high probability while only requiring an expected $O(\frac{1}{ε^2}\log^2 \frac{n}{ε^2})$ comparisons on average per voter. Moreover, this voting protocol is shown to have a quasi-truthful Nash equilibrium: namely, a Nash equilibrium exists which may not be truthful, but produces a winner with the same probability distribution as that of the truthful strategy.

cs.GT↗

arXiv2012

Biased Assimilation, Homophily and the Dynamics of Polarization

Are we as a society getting more polarized, and if so, why? We try to answer this question through a model of opinion formation. Empirical studies have shown that homophily results in polarization. However, we show that DeGroot's well-known model of opinion formation based on repeated averaging can never be polarizing, even if individuals are arbitrarily homophilous. We generalize DeGroot's model to account for a phenomenon well-known in social psychology as biased assimilation: when presented with mixed or inconclusive evidence on a complex issue, individuals draw undue support for their initial position thereby arriving at a more extreme opinion. We show that in a simple model of homophilous networks, our biased opinion formation process results in either polarization, persistent disagreement or consensus depending on how biased individuals are. In other words, homophily alone, without biased assimilation, is not sufficient to polarize society. Quite interestingly, biased assimilation also provides insight into the following related question: do internet based recommender algorithms that show us personalized content contribute to polarization? We make a connection between biased assimilation and the polarizing effects of some random-walk based recommender algorithms that are similar in spirit to some commonly used recommender algorithms.

cs.SI↗

arXiv2012

Single pass sparsification in the streaming model with edge deletions

In this paper we give a construction of cut sparsifiers of Benczur and Karger in the {\em dynamic} streaming setting in a single pass over the data stream. Previous constructions either required multiple passes or were unable to handle edge deletions. We use $\tilde{O}(1/\e^2)$ time for each stream update and $\tilde{O}(n/\e^2)$ time to construct a sparsifier. Our $\e$-sparsifiers have $O(n\log^3 n/\e^2)$ edges. The main tools behind our result are an application of sketching techniques of Ahn et al.[SODA'12] to estimate edge connectivity together with a novel application of sampling with limited independence and sparse recovery to produce the edges of the sparsifier.

cs.DS↗

arXiv2012

Liquidity in Credit Networks: A Little Trust Goes a Long Way

Credit networks represent a way of modeling trust between entities in a network. Nodes in the network print their own currency and trust each other for a certain amount of each other's currency. This allows the network to serve as a decentralized payment infrastructure---arbitrary payments can be routed through the network by passing IOUs between trusting nodes in their respective currencies---and obviates the need for a common currency. Nodes can repeatedly transact with each other and pay for the transaction using trusted currency. A natural question to ask in this setting is: how long can the network sustain liquidity, i.e., how long can the network support the routing of payments before credit dries up? We answer this question in terms of the long term failure probability of transactions for various network topologies and credit values. We prove that the transaction failure probability is independent of the path along which transactions are routed. We show that under symmetric transaction rates, the transaction failure probability in a number of well-known graph families goes to zero as the size, density or credit capacity of the network increases. We also show via simulations that even networks of small size and credit capacity can route transactions with high probability if they are well-connected. Further, we characterize a centralized currency system as a special type of a star network (one where edges to the root have infinite credit capacity, and transactions occur only between leaf nodes) and compute the steady-state transaction failure probability in a centralized system. We show that liquidity in star networks, complete graphs and Erdös-Rényi networks is comparable to that in equivalent centralized currency systems; thus we do not lose much liquidity in return for their robustness and decentralized properties.

q-fin.TR↗

arXiv2011

Widescope - A social platform for serious conversations on the Web

There are several web platforms that people use to interact and exchange ideas, such as social networks like Facebook, Twitter, and Google+; Q&A sites like Quora and Yahoo! Answers; and myriad independent fora. However, there is a scarcity of platforms that facilitate discussion of complex subjects where people with divergent views can easily rationalize their points of view using a shared knowledge base, and leverage it towards shared objectives, e.g. to arrive at a mutually acceptable compromise. In this paper, as a first step, we present Widescope, a novel collaborative web platform for catalyzing shared understanding of the US Federal and State budget debates in order to help users reach data-driven consensus about the complex issues involved. It aggregates disparate sources of financial data from different budgets (i.e. from past, present, and proposed) and presents a unified interface using interactive visualizations. It leverages distributed collaboration to encourage exploration of ideas and debate. Users can propose budgets ab-initio, support existing proposals, compare between different budgets, and collaborate with others in real time. We hypothesize that such a platform can be useful in bringing people's thoughts and opinions closer. Toward this, we present preliminary evidence from a simple pilot experiment, using triadic voting (which we also formally analyze to show that is better than hot-or-not voting), that 5 out of 6 groups of users with divergent views (conservatives vs liberals) come to a consensus while aiming to halve the deficit using Widescope. We believe that tools like Widescope could have a positive impact on other complex, data-driven social issues.

cs.SI↗

arXiv2010

Perfect Matchings in O(n \log n) Time in Regular Bipartite Graphs

In this paper we consider the well-studied problem of finding a perfect matching in a d-regular bipartite graph on 2n nodes with m=nd edges. The best-known algorithm for general bipartite graphs (due to Hopcroft and Karp) takes time O(m\sqrt{n}). In regular bipartite graphs, however, a matching is known to be computable in O(m) time (due to Cole, Ost and Schirra). In a recent line of work by Goel, Kapralov and Khanna the O(m) time algorithm was improved first to \tilde O(min{m, n^{2.5}/d}) and then to \tilde O(min{m, n^2/d}). It was also shown that the latter algorithm is optimal up to polylogarithmic factors among all algorithms that use non-adaptive uniform sampling to reduce the size of the graph as a first step. In this paper, we give a randomized algorithm that finds a perfect matching in a d-regular graph and runs in O(n\log n) time (both in expectation and with high probability). The algorithm performs an appropriately truncated random walk on a modified graph to successively find augmenting paths. Our algorithm may be viewed as using adaptive uniform sampling, and is thus able to bypass the limitations of (non-adaptive) uniform sampling established in earlier work. We also show that randomization is crucial for obtaining o(nd) time algorithms by establishing an Ω(nd) lower bound for any deterministic algorithm. Our techniques also give an algorithm that successively finds a matching in the support of a doubly stochastic matrix in expected time O(n\log^2 n) time, with O(m) pre-processing time; this gives a simple O(m+mn\log^2 n) time algorithm for finding the Birkhoff-von Neumann decomposition of a doubly stochastic matrix.

cs.DS↗

arXiv2010

Fast Incremental and Personalized PageRank

In this paper, we analyze the efficiency of Monte Carlo methods for incremental computation of PageRank, personalized PageRank, and similar random walk based methods (with focus on SALSA), on large-scale dynamically evolving social networks. We assume that the graph of friendships is stored in distributed shared memory, as is the case for large social networks such as Twitter. For global PageRank, we assume that the social network has $n$ nodes, and $m$ adversarially chosen edges arrive in a random order. We show that with a reset probability of $ε$, the total work needed to maintain an accurate estimate (using the Monte Carlo method) of the PageRank of every node at all times is $O(\frac{n\ln m}{ε^{2}})$. This is significantly better than all known bounds for incremental PageRank. For instance, if we naively recompute the PageRanks as each edge arrives, the simple power iteration method needs $Ω(\frac{m^2}{\ln(1/(1-ε))})$ total time and the Monte Carlo method needs $O(mn/ε)$ total time; both are prohibitively expensive. Furthermore, we also show that we can handle deletions equally efficiently. We then study the computation of the top $k$ personalized PageRanks starting from a seed node, assuming that personalized PageRanks follow a power-law with exponent $α< 1$. We show that if we store $R>q\ln n$ random walks starting from every node for large enough constant $q$ (using the approach outlined for global PageRank), then the expected number of calls made to the distributed social network database is $O(k/(R^{(1-α)/α}))$. We also present experimental results from the social networking site, Twitter, verifying our assumptions and analyses. The overall result is that this algorithm is fast enough for real-time queries over a dynamic social network.

cs.DS↗

arXiv2010

One Tree Suffices: A Simultaneous O(1)-Approximation for Single-Sink Buy-at-Bulk

We study the single-sink buy-at-bulk problem with an unknown cost function. We wish to route flow from a set of demand nodes to a root node, where the cost of routing x total flow along an edge is proportional to f(x) for some concave, non-decreasing function f satisfying f(0)=0. We present a simple, fast, combinatorial algorithm that takes a set of demands and constructs a single tree T such that for all f the cost f(T) is a 47.45-approximation of the optimal cost for that f. This is within a factor of 2.33 of the best approximation ratio currently achievable when the tree can be optimized for a specific function. Trees achieving simultaneous O(1)-approximations for all concave functions were previously not known to exist regardless of computation time.

cs.DS↗

arXiv2010

Similarity Search and Locality Sensitive Hashing using TCAMs

Similarity search methods are widely used as kernels in various machine learning applications. Nearest neighbor search (NNS) algorithms are often used to retrieve similar entries, given a query. While there exist efficient techniques for exact query lookup using hashing, similarity search using exact nearest neighbors is known to be a hard problem and in high dimensions, best known solutions offer little improvement over a linear scan. Fast solutions to the approximate NNS problem include Locality Sensitive Hashing (LSH) based techniques, which need storage polynomial in $n$ with exponent greater than $1$, and query time sublinear, but still polynomial in $n$, where $n$ is the size of the database. In this work we present a new technique of solving the approximate NNS problem in Euclidean space using a Ternary Content Addressable Memory (TCAM), which needs near linear space and has O(1) query time. In fact, this method also works around the best known lower bounds in the cell probe model for the query time using a data structure near linear in the size of the data base. TCAMs are high performance associative memories widely used in networking applications such as access control lists. A TCAM can query for a bit vector within a database of ternary vectors, where every bit position represents $0$, $1$ or $*$. The $*$ is a wild card representing either a $0$ or a $1$. We leverage TCAMs to design a variant of LSH, called Ternary Locality Sensitive Hashing (TLSH) wherein we hash database entries represented by vectors in the Euclidean space into $\{0,1,*\}$. By using the added functionality of a TLSH scheme with respect to the $*$ character, we solve an instance of the approximate nearest neighbor problem with 1 TCAM access and storage nearly linear in the size of the database. We believe that this work can open new avenues in very high speed data mining.

cs.DB↗

arXiv2010

Graph Sparsification via Refinement Sampling

A graph G'(V,E') is an \eps-sparsification of G for some \eps>0, if every (weighted) cut in G' is within (1\pm \eps) of the corresponding cut in G. A celebrated result of Benczur and Karger shows that for every undirected graph G, an \eps-sparsification with O(n\log n/\e^2) edges can be constructed in O(m\log^2n) time. Applications to modern massive data sets often constrain algorithms to use computation models that restrict random access to the input. The semi-streaming model, in which the algorithm is constrained to use \tilde O(n) space, has been shown to be a good abstraction for analyzing graph algorithms in applications to large data sets. Recently, a semi-streaming algorithm for graph sparsification was presented by Anh and Guha; the total running time of their implementation is Ω(mn), too large for applications where both space and time are important. In this paper, we introduce a new technique for graph sparsification, namely refinement sampling, that gives an \tilde{O}(m) time semi-streaming algorithm for graph sparsification. Specifically, we show that refinement sampling can be used to design a one-pass streaming algorithm for sparsification that takes O(\log\log n) time per edge, uses O(\log^2 n) space per node, and outputs an \eps-sparsifier with O(n\log^3 n/\eps^2) edges. At a slightly increased space and time complexity, we can reduce the sparsifier size to O(n \log n/\e^2) edges matching the Benczur-Karger result, while improving upon the Benczur-Karger runtime for m=ω(n\log^3 n). Finally, we show that an \eps-sparsifier with O(n \log n/\eps^2) edges can be constructed in two passes over the data and O(m) time whenever m =Ω(n^{1+δ}) for some constant δ>0. As a by-product of our approach, we also obtain an O(m\log\log n+n \log n) time streaming algorithm to compute a sparse k-connectivity certificate of a graph.

cs.DS↗

arXiv2009

An Oblivious O(1)-Approximation for Single Source Buy-at-Bulk

We consider the single-source (or single-sink) buy-at-bulk problem with an unknown concave cost function. We want to route a set of demands along a graph to or from a designated root node, and the cost of routing x units of flow along an edge is proportional to some concave, non-decreasing function f such that f(0) = 0. We present a polynomial time algorithm that finds a distribution over trees such that the expected cost of a tree for any f is within an O(1)-factor of the optimum cost for that f. The previous best simultaneous approximation for this problem, even ignoring computation time, was O(log |D|), where D is the multi-set of demand nodes. We design a simple algorithmic framework using the ellipsoid method that finds an O(1)-approximation if one exists, and then construct a separation oracle using a novel adaptation of the Guha, Meyerson, and Munagala algorithm for the single-sink buy-at-bulk problem that proves an O(1) approximation is possible for all f. The number of trees in the support of the distribution constructed by our algorithm is at most 1+log |D|.

cs.DS↗

arXiv2009

Perfect Matchings in Õ(n^{1.5}) Time in Regular Bipartite Graphs

We consider the well-studied problem of finding a perfect matching in $d$-regular bipartite graphs with $2n$ vertices and $m = nd$ edges. While the best-known algorithm for general bipartite graphs (due to Hopcroft and Karp) takes $O(m \sqrt{n})$ time, in regular bipartite graphs, a perfect matching is known to be computable in $O(m)$ time. Very recently, the $O(m)$ bound was improved to $O(\min\{m, \frac{n^{2.5}\ln n}{d}\})$ expected time, an expression that is bounded by $\tilde{O}(n^{1.75})$. In this paper, we further improve this result by giving an $O(\min\{m, \frac{n^2\ln^3 n}{d}\})$ expected time algorithm for finding a perfect matching in regular bipartite graphs; as a function of $n$ alone, the algorithm takes expected time $O((n\ln n)^{1.5})$. To obtain this result, we design and analyze a two-stage sampling scheme that reduces the problem of finding a perfect matching in a regular bipartite graph to the same problem on a subsampled bipartite graph with $O(n\ln n)$ edges that has a perfect matching with high probability. The matching is then recovered using the Hopcroft-Karp algorithm. While the standard analysis of Hopcroft-Karp gives us an $\tilde{O}(n^{1.5})$ running time, we present a tighter analysis for our special case that results in the stronger $\tilde{O}(\min\{m, \frac{n^2}{d} \})$ time mentioned earlier. Our proof of correctness of this sampling scheme uses a new correspondence theorem between cuts and Hall's theorem ``witnesses'' for a perfect matching in a bipartite graph that we prove. We believe this theorem may be of independent interest; as another example application, we show that a perfect matching in the support of an $n \times n$ doubly stochastic matrix with $m$ non-zero entries can be found in expected time $\tilde{O}(m + n^{1.5})$.

cs.DS↗

arXiv2009

Hybrid Keyword Search Auctions

Search auctions have become a dominant source of revenue generation on the Internet. Such auctions have typically used per-click bidding and pricing. We propose the use of hybrid auctions where an advertiser can make a per-impression as well as a per-click bid, and the auctioneer then chooses one of the two as the pricing mechanism. We assume that the advertiser and the auctioneer both have separate beliefs (called priors) on the click-probability of an advertisement. We first prove that the hybrid auction is truthful, assuming that the advertisers are risk-neutral. We then show that this auction is superior to the existing per-click auction in multiple ways: 1) It takes into account the risk characteristics of the advertisers. 2) For obscure keywords, the auctioneer is unlikely to have a very sharp prior on the click-probabilities. In such situations, the hybrid auction can result in significantly higher revenue. 3) An advertiser who believes that its click-probability is much higher than the auctioneer's estimate can use per-impression bids to correct the auctioneer's prior without incurring any extra cost. 4) The hybrid auction can allow the advertiser and auctioneer to implement complex dynamic programming strategies. As Internet commerce matures, we need more sophisticated pricing models to exploit all the information held by each of the participants. We believe that hybrid auctions could be an important step in this direction.

cs.GT↗