arXiv subjects

Richard Combes

Publications and source records attributed to Richard Combes.

At least 37 records · Page 2Linked to original sources

arXiv2021

Asymptotically Optimal Strategies For Combinatorial Semi-Bandits in Polynomial Time

We consider combinatorial semi-bandits with uncorrelated Gaussian rewards. In this article, we propose the first method, to the best of our knowledge, that enables to compute the solution of the Graves-Lai optimization problem in polynomial time for many combinatorial structures of interest. In turn, this immediately yields the first known approach to implement asymptotically optimal algorithms in polynomial time for combinatorial semi-bandits.

stat.ML↗

arXiv2021

Statistically Efficient, Polynomial Time Algorithms for Combinatorial Semi Bandits

We consider combinatorial semi-bandits over a set of arms ${\cal X} \subset \{0,1\}^d$ where rewards are uncorrelated across items. For this problem, the algorithm ESCB yields the smallest known regret bound $R(T) = {\cal O}\Big( {d (\ln m)^2 (\ln T) \over Δ_{\min} }\Big)$, but it has computational complexity ${\cal O}(|{\cal X}|)$ which is typically exponential in $d$, and cannot be used in large dimensions. We propose the first algorithm which is both computationally and statistically efficient for this problem with regret $R(T) = {\cal O} \Big({d (\ln m)^2 (\ln T)\over Δ_{\min} }\Big)$ and computational complexity ${\cal O}(T {\bf poly}(d))$. Our approach involves carefully designing an approximate version of ESCB with the same regret guarantees, showing that this approximate algorithm can be implemented in time ${\cal O}(T {\bf poly}(d))$ by repeatedly maximizing a linear function over ${\cal X}$ subject to a linear budget constraint, and showing how to solve this maximization problems efficiently.

stat.ML↗

arXiv2020

Solving Random Parity Games in Polynomial Time

We consider the problem of solving random parity games. We prove that parity games exibit a phase transition threshold above $d_P$, so that when the degree of the graph that defines the game has a degree $d > d_P$ then there exists a polynomial time algorithm that solves the game with high probability when the number of nodes goes to infinity. We further propose the SWCP (Self-Winning Cycles Propagation) algorithm and show that, when the degree is large enough, SWCP solves the game with high probability. Furthermore, the complexity of SWCP is polynomial $O\Big(|{\cal V}|^2 + |{\cal V}||{\cal E}|\Big)$. The design of SWCP is based on the threshold for the appearance of particular types of cycles in the players' respective subgraphs. We further show that non-sparse games can be solved in time $O(|{\cal V}|)$ with high probability, and emit a conjecture concerning the hardness of the $d=2$ case.

cs.LO↗

arXiv2019

Solving Bernoulli Rank-One Bandits with Unimodal Thompson Sampling

Stochastic Rank-One Bandits (Katarya et al, (2017a,b)) are a simple framework for regret minimization problems over rank-one matrices of arms. The initially proposed algorithms are proved to have logarithmic regret, but do not match the existing lower bound for this problem. We close this gap by first proving that rank-one bandits are a particular instance of unimodal bandits, and then providing a new analysis of Unimodal Thompson Sampling (UTS), initially proposed by Paladino et al (2017). We prove an asymptotically optimal regret bound on the frequentist regret of UTS and we support our claims with simulations showing the significant improvement of our method compared to the state-of-the-art.

stat.ML↗

arXiv2019

Computationally Efficient Estimation of the Spectral Gap of a Markov Chain

We consider the problem of estimating from sample paths the absolute spectral gap $γ_*$ of a reversible, irreducible and aperiodic Markov chain $(X_t)_{t \in \mathbb{N}}$ over a finite state space $Ω$. We propose the ${\tt UCPI}$ (Upper Confidence Power Iteration) algorithm for this problem, a low-complexity algorithm which estimates the spectral gap in time ${\cal O}(n)$ and memory space ${\cal O}((\ln n)^2)$ given $n$ samples. This is in stark contrast with most known methods which require at least memory space ${\cal O}(|Ω|)$, so that they cannot be applied to large state spaces. Furthermore, ${\tt UCPI}$ is amenable to parallel implementation.

stat.ML↗

arXiv2018

Hierarchical Beamforming: Resource Allocation, Fairness and Flow Level Performance

We consider hierarchical beamforming in wireless networks. For a given population of flows, we propose computationally efficient algorithms for fair rate allocation including proportional fairness and max-min fairness. We next propose closed-form formulas for flow level performance, for both elastic (with either proportional fairness and max-min fairness) and streaming traffic. We further assess the performance of hierarchical beamforming using numerical experiments. Since the proposed solutions have low complexity compared to conventional beamforming, our work suggests that hierarchical beamforming is a promising candidate for the implementation of beamforming in future cellular networks.

cs.NI↗

arXiv2018

Device-to-Device Aided Multicasting

We consider a device-to-device (D2D) aided multicast channel, where a transmitter wishes to convey a common message to many receivers and these receivers cooperate with each other. We propose a simple computationally efficient scheme requiring only statistical channel knowledge at transmitter. Our analysis in general topologies reveals that, when the number of receivers $K$ grows to infinity, the proposed scheme guarantees a multicast rate of ${1 \over 2} \log_2(1 + β\ln K )$ with high probability for any $β< β^\star$ where $β^\star$ depends on the network topology. This scheme undergoes a phase transition at threshold $β^\star \ln K$ where transmissions are successful/unsuccessful with high probability when the SNR is above/below this threshold. We also analyze the outage rate of the proposed scheme in the same setting.

cs.IT↗

arXiv2018

Utility Optimal Scheduling for Coded Caching in General Topologies

We consider coded caching over the fading broadcast channel, where the users, equipped with a memory of finite size, experience asymmetric fading statistics. It is known that a naive application of coded caching over the channel at hand performs poorly especially in the regime of a large number of users due to a vanishing multicast rate. We overcome this detrimental effect by a careful design of opportunistic scheduling policies such that some utility function of the long-term average rates should be maximized while balancing fairness among users. In particular, we propose a threshold-based scheduling that requires only statistical channel state information and one-bit feedback from each user. More specifically, each user indicates via feedback whenever its SNR is above a threshold determined solely by the fading statistics and the fairness requirement. Surprisingly, we prove that this simple scheme achieves the optimal utility in the regime of a large number of users. Numerical examples show that our proposed scheme performs closely to the scheduling with full channel state information, but at a significantly reduced complexity.

cs.IT↗

arXiv2017

Minimal Exploration in Structured Stochastic Bandits

This paper introduces and addresses a wide class of stochastic bandit problems where the function mapping the arm to the corresponding reward exhibits some known structural properties. Most existing structures (e.g. linear, Lipschitz, unimodal, combinatorial, dueling, ...) are covered by our framework. We derive an asymptotic instance-specific regret lower bound for these problems, and develop OSSB, an algorithm whose regret matches this fundamental limit. OSSB is not based on the classical principle of "optimism in the face of uncertainty" or on Thompson sampling, and rather aims at matching the minimal exploration rates of sub-optimal arms as characterized in the derivation of the regret lower bound. We illustrate the efficiency of OSSB using numerical experiments in the case of the linear bandit problem and show that OSSB outperforms existing algorithms, including Thompson sampling.

stat.ML↗

arXiv2017

A Minimax Optimal Algorithm for Crowdsourcing

We consider the problem of accurately estimating the reliability of workers based on noisy labels they provide, which is a fundamental question in crowdsourcing. We propose a novel lower bound on the minimax estimation error which applies to any estimation procedure. We further propose Triangular Estimation (TE), an algorithm for estimating the reliability of workers. TE has low complexity, may be implemented in a streaming setting when labels are provided by workers in real time, and does not rely on an iterative procedure. We further prove that TE is minimax optimal and matches our lower bound. We conclude by assessing the performance of TE and other state-of-the-art algorithms on both synthetic and real-world data sets.

stat.ML↗

arXiv2017

An Approximate ML Detector for MIMO Channels Corrupted by Phase Noise

We consider the multiple-input multiple-output (MIMO) communication channel impaired by phase noises at both the transmitter and receiver. We focus on the maximum likelihood (ML) detection problem for uncoded single-carrier transmission. We derive an approximation of the likelihood function, based on which we propose an efficient detection algorithm. The proposed algorithm, named self-interference whitening (SIW), consists in 1) estimating the self-interference caused by the phase noise perturbation, then 2) whitening the said interference, and finally 3) detecting the transmitted vector. While the exact ML solution is computationally intractable, we construct a simulation-based lower bound on the error probability of ML detection. Leveraging this lower bound, we perform extensive numerical experiments demonstrating that SIW is, in most cases of interest, very close to optimal with moderate phase noise. More importantly and perhaps surprisingly, such near-ML performance can be achieved by applying only twice the nearest neighbor detection algorithm. In this sense, our results reveal a striking fact: near-ML detection of phase noise corrupted MIMO channels can be done as efficiently as for conventional MIMO channels without phase noise.

cs.IT↗

arXiv2017

Opportunistic Content Delivery in Fading Broadcast Channels

We consider content delivery over fading broadcast channels. A server wants to transmit K files to K users, each equipped with a cache of finite size. Using the coded caching scheme of Maddah-Ali and Niesen, we design an opportunistic delivery scheme where the long-term sum content delivery rate scales with K the number of users in the system. The proposed delivery scheme combines superposition coding together with appropriate power allocation across sub-files intended to different subsets of users. We analyze the long-term average sum content delivery rate achieved by two special cases of our scheme: a) a selection scheme that chooses the subset of users with the largest weighted rate, and b) a baseline scheme that transmits to K users using the scheme of Maddah-Ali and Niesen. We prove that coded caching with appropriate user selection is scalable since it yields a linear increase of the average sum content delivery rate.

cs.IT↗

arXiv2017

Stochastic Online Shortest Path Routing: The Value of Feedback

This paper studies online shortest path routing over multi-hop networks. Link costs or delays are time-varying and modeled by independent and identically distributed random processes, whose parameters are initially unknown. The parameters, and hence the optimal path, can only be estimated by routing packets through the network and observing the realized delays. Our aim is to find a routing policy that minimizes the regret (the cumulative difference of expected delay) between the path chosen by the policy and the unknown optimal path. We formulate the problem as a combinatorial bandit optimization problem and consider several scenarios that differ in where routing decisions are made and in the information available when making the decisions. For each scenario, we derive a tight asymptotic lower bound on the regret that has to be satisfied by any online routing policy. These bounds help us to understand the performance improvements we can expect when (i) taking routing decisions at each hop rather than at the source only, and (ii) observing per-link delays rather than end-to-end path delays. In particular, we show that (i) is of no use while (ii) can have a spectacular impact. Three algorithms, with a trade-off between computational complexity and performance, are proposed. The regret upper bounds of these algorithms improve over those of the existing algorithms, and they significantly outperform state-of-the-art algorithms in numerical experiments.

cs.NI↗

arXiv2016

Multipath streaming: fundamental limits and efficient algorithms

We investigate streaming over multiple links. A file is split into small units called chunks that may be requested on the various links according to some policy, and received after some random delay. After a start-up time called pre-buffering time, received chunks are played at a fixed speed. There is starvation if the chunk to be played has not yet arrived. We provide lower bounds (fundamental limits) on the starvation probability of any policy. We further propose simple, order-optimal policies that require no feedback. For general delay distributions, we provide tractable upper bounds for the starvation probability of the proposed policies, allowing to select the pre-buffering time appropriately. We specialize our results to: (i) links that employ CSMA or opportunistic scheduling at the packet level, (ii) links shared with a primary user (iii) links that use fair rate sharing at the flow level. We consider a generic model so that our results give insight into the design and performance of media streaming over (a) wired networks with several paths between the source and destination, (b) wireless networks featuring spectrum aggregation and (c) multi-homed wireless networks.

cs.NI↗

arXiv2016

A Streaming Algorithm for Crowdsourced Data Classification

We propose a streaming algorithm for the binary classification of data based on crowdsourcing. The algorithm learns the competence of each labeller by comparing her labels to those of other labellers on the same tasks and uses this information to minimize the prediction error rate on each task. We provide performance guarantees of our algorithm for a fixed population of independent labellers. In particular, we show that our algorithm is optimal in the sense that the cumulative regret compared to the optimal decision with known labeller error probabilities is finite, independently of the number of tasks to label. The complexity of the algorithm is linear in the number of labellers and the number of tasks, up to some logarithmic factors. Numerical experiments illustrate the performance of our algorithm compared to existing algorithms, including simple majority voting and expectation-maximization algorithms, on both synthetic and real datasets.

stat.ML↗

arXiv2015

Combinatorial Bandits Revisited

This paper investigates stochastic and adversarial combinatorial multi-armed bandit problems. In the stochastic setting under semi-bandit feedback, we derive a problem-specific regret lower bound, and discuss its scaling with the dimension of the decision space. We propose ESCB, an algorithm that efficiently exploits the structure of the problem and provide a finite-time analysis of its regret. ESCB has better performance guarantees than existing algorithms, and significantly outperforms these algorithms in practice. In the adversarial setting under bandit feedback, we propose \textsc{CombEXP}, an algorithm with the same regret scaling as state-of-the-art algorithms, but with lower computational complexity for some combinatorial problems.

cs.LG↗

arXiv2015

Unimodal Bandits without Smoothness

We consider stochastic bandit problems with a continuous set of arms and where the expected reward is a continuous and unimodal function of the arm. No further assumption is made regarding the smoothness and the structure of the expected reward function. For these problems, we propose the Stochastic Pentachotomy (SP) algorithm, and derive finite-time upper bounds on its regret and optimization error. In particular, we show that, for any expected reward function $μ$ that behaves as $μ(x)=μ(x^\star)-C|x-x^\star|^ξ$ locally around its maximizer $x^\star$ for some $ξ, C>0$, the SP algorithm is order-optimal. Namely its regret and optimization error scale as $O(\sqrt{T\log(T)})$ and $O(\sqrt{\log(T)/T})$, respectively, when the time horizon $T$ grows large. These scalings are achieved without the knowledge of $ξ$ and $C$. Our algorithm is based on asymptotically optimal sequential statistical tests used to successively trim an interval that contains the best arm with high probability. To our knowledge, the SP algorithm constitutes the first sequential arm selection rule that achieves a regret and optimization error scaling as $O(\sqrt{T})$ and $O(1/\sqrt{T})$, respectively, up to a logarithmic factor for non-smooth expected reward functions, as well as for smooth functions with unknown smoothness.

cs.LG↗

arXiv2014

Unimodal Bandits: Regret Lower Bounds and Optimal Algorithms

We consider stochastic multi-armed bandits where the expected reward is a unimodal function over partially ordered arms. This important class of problems has been recently investigated in (Cope 2009, Yu 2011). The set of arms is either discrete, in which case arms correspond to the vertices of a finite graph whose structure represents similarity in rewards, or continuous, in which case arms belong to a bounded interval. For discrete unimodal bandits, we derive asymptotic lower bounds for the regret achieved under any algorithm, and propose OSUB, an algorithm whose regret matches this lower bound. Our algorithm optimally exploits the unimodal structure of the problem, and surprisingly, its asymptotic regret does not depend on the number of arms. We also provide a regret upper bound for OSUB in non-stationary environments where the expected rewards smoothly evolve over time. The analytical results are supported by numerical experiments showing that OSUB performs significantly better than the state-of-the-art algorithms. For continuous sets of arms, we provide a brief discussion. We show that combining an appropriate discretization of the set of arms with the UCB algorithm yields an order-optimal regret, and in practice, outperforms recently proposed algorithms designed to exploit the unimodal structure.

cs.LG↗