Searcharxiv⌕ Search

arXiv subjects

Sinong Wang

Publications and source records attributed to Sinong Wang.

45 records · Page 3Linked to original sources
arXiv2018

Fundamental Limits of Coded Linear Transform

In large scale distributed linear transform problems, coded computation plays an important role to effectively deal with "stragglers" (distributed computations that may get delayed due to few slow or faulty processors). We propose a coded computation strategy, referred to as diagonal code, that achieves the optimum recovery threshold and the optimum computation load. This is the first code that simultaneously achieves two-fold optimality in coded distributed linear transforms. Furthermore, by leveraging the idea from random proposal graph theory, we design two random codes that can guarantee optimum recovery threshold with high probability but with much less computation load. These codes provide order-wise improvement over the state-of-the-art. Moreover, the experimental results show significant improvement compared to both uncoded and existing coding schemes.

cs.IT↗
arXiv2018

Coded Sparse Matrix Multiplication

In a large-scale and distributed matrix multiplication problem $C=A^{\intercal}B$, where $C\in\mathbb{R}^{r\times t}$, the coded computation plays an important role to effectively deal with "stragglers" (distributed computations that may get delayed due to few slow or faulty processors). However, existing coded schemes could destroy the significant sparsity that exists in large-scale machine learning problems, and could result in much higher computation overhead, i.e., $O(rt)$ decoding time. In this paper, we develop a new coded computation strategy, we call \emph{sparse code}, which achieves near \emph{optimal recovery threshold}, \emph{low computation overhead}, and \emph{linear decoding time} $O(nnz(C))$. We implement our scheme and demonstrate the advantage of the approach over both uncoded and current fastest coded strategies.

cs.DC↗
arXiv2018

UCBoost: A Boosting Approach to Tame Complexity and Optimality for Stochastic Bandits

In this work, we address the open problem of finding low-complexity near-optimal multi-armed bandit algorithms for sequential decision making problems. Existing bandit algorithms are either sub-optimal and computationally simple (e.g., UCB1) or optimal and computationally complex (e.g., kl-UCB). We propose a boosting approach to Upper Confidence Bound based algorithms for stochastic bandits, that we call UCBoost. Specifically, we propose two types of UCBoost algorithms. We show that UCBoost($D$) enjoys $O(1)$ complexity for each arm per round as well as regret guarantee that is $1/e$-close to that of the kl-UCB algorithm. We propose an approximation-based UCBoost algorithm, UCBoost($ε$), that enjoys a regret guarantee $ε$-close to that of kl-UCB as well as $O(\log(1/ε))$ complexity for each arm per round. Hence, our algorithms provide practitioners a practical way to trade optimality with computational complexity. Finally, we present numerical results which show that UCBoost($ε$) can achieve the same regret performance as the standard kl-UCB while incurring only $1\%$ of the computational cost of kl-UCB.

cs.LG↗
arXiv2017

Towards Fast-Convergence, Low-Delay and Low-Complexity Network Optimization

Distributed network optimization has been studied for well over a decade. However, we still do not have a good idea of how to design schemes that can simultaneously provide good performance across the dimensions of utility optimality, convergence speed, and delay. To address these challenges, in this paper, we propose a new algorithmic framework with all these metrics approaching optimality. The salient features of our new algorithm are three-fold: (i) fast convergence: it converges with only $O(\log(1/ε))$ iterations that is the fastest speed among all the existing algorithms; (ii) low delay: it guarantees optimal utility with finite queue length; (iii) simple implementation: the control variables of this algorithm are based on virtual queues that do not require maintaining per-flow information. The new technique builds on a kind of inexact Uzawa method in the Alternating Directional Method of Multiplier, and provides a new theoretical path to prove global and linear convergence rate of such a method without requiring the full rank assumption of the constraint matrix.

cs.NI↗
arXiv2017

Security Game with Non-additive Utilities and Multiple Attacker Resources

There has been significant interest in studying security games for modeling the interplay of attacks and defenses on various systems involving critical infrastructure, financial system security, political campaigns, and civil safeguarding. However, existing security game models typically either assume additive utility functions, or that the attacker can attack only one target. Such assumptions lead to tractable analysis, but miss key inherent dependencies that exist among different targets in current complex networks. In this paper, we generalize the classical security game models to allow for non-additive utility functions. We also allow attackers to be able to attack multiple targets. We examine such a general security game from a theoretical perspective and provide a unified view. In particular, we show that each security game is equivalent to a combinatorial optimization problem over a set system $\varepsilon$, which consists of defender's pure strategy space. The key technique we use is based on the transformation, projection of a polytope, and the elipsoid method. This work settles several open questions in security game domain and significantly extends the state of-the-art of both the polynomial solvable and NP-hard class of the security game.

cs.GT↗
arXiv2015

The Performance Analysis of Coded Cache in Wireless Fading Channel

The rapid growth of data volume and the accompanying congestion problems over the wireless networks have been critical issues to content providers. A novel technique, termed as coded cache, is proposed to relieve the burden. Through creating coded-multicasting opportunities, the coded-cache scheme can provide extra performance gain over the conventional push technique that simply pre-stores contents at local caches during the network idle period. But existing works on the coded caching scheme assumed the availability of an error-free shared channel accessible by each user. This paper considers the more realistic scenario where each user may experience different link quality. In this case, the system performance would be restricted by the user with the worst channel condition. And the corresponding resource allocation schemes aimed at breaking this obstacles are developed. Specifically, we employ the coded caching scheme in time division and frequency division transmission mode and formulate the sub-optimal problems. Power and bandwidth are allocated respectively to maximum the system throughput. The simulation results show that the throughput of the technique in wireless scenario will be limited and would decrease as the number of users becomes sufficiently large.

cs.NI↗
arXiv2014

Exploiting the Unexploited of Coded Caching for Wireless Content Distribution: Detailed Theoretical Proofs

Recent studies show that the coded caching technique can facilitate the wireless content distribution by mitigating the wireless traffic rate during the peak-traffic time, where the contents are partially prefetched to the local cache of mobile devices during the off-peak time. The remaining contents are then jointly coded and delivered in multicast, when many content requests are initiated in the peak-traffic time. The requested contents can be recovered from the local-prefetched and multicast data with requesters experiencing less congestions. However, the benefit of the coded caching scheme is still under estimated, where the potential gain by appropriate caching distribution is under exploited. In this paper, we propose a theoretical model to minimize the average wireless traffic rate required in the coded caching, for which the optimized caching distribution is derived with the content popularity distribution taken into account. In order to improve the computational efficiency for determining the appropriate caching distribution, we transform the objective function from the average wireless traffic rate into the average size of un-prefetched contents. We theoretically show the order optimality of the derived results from both the primal model and the relaxed one. Simulation results show that the coded caching performance can be further improved with the derived caching distribution.

cs.NI↗