arXiv subjects

András György

Publications and source records attributed to András György.

50 records · Page 3Linked to original sources

arXiv2018

A Weakness Measure for GR(1) Formulae

In spite of the theoretical and algorithmic developments for system synthesis in recent years, little effort has been dedicated to quantifying the quality of the specifications used for synthesis. When dealing with unrealizable specifications, finding the weakest environment assumptions that would ensure realizability is typically a desirable property; in such context the weakness of the assumptions is a major quality parameter. The question of whether one assumption is weaker than another is commonly interpreted using implication or, equivalently, language inclusion. However, this interpretation does not provide any further insight into the weakness of assumptions when implication does not hold. To our knowledge, the only measure that is capable of comparing two formulae in this case is entropy, but even it fails to provide a sufficiently refined notion of weakness in case of GR(1) formulae, a subset of linear temporal logic formulae which is of particular interest in controller synthesis. In this paper we propose a more refined measure of weakness based on the Hausdorff dimension, a concept that captures the notion of size of the omega-language satisfying a linear temporal logic formula. We identify the conditions under which this measure is guaranteed to distinguish between weaker and stronger GR(1) formulae. We evaluate our proposed weakness measure in the context of computing GR(1) assumptions refinements.

cs.LO↗

arXiv2017

A Modular Analysis of Adaptive (Non-)Convex Optimization: Optimism, Composite Objectives, and Variational Bounds

Recently, much work has been done on extending the scope of online learning and incremental stochastic optimization algorithms. In this paper we contribute to this effort in two ways: First, based on a new regret decomposition and a generalization of Bregman divergences, we provide a self-contained, modular analysis of the two workhorses of online learning: (general) adaptive versions of Mirror Descent (MD) and the Follow-the-Regularized-Leader (FTRL) algorithms. The analysis is done with extra care so as not to introduce assumptions not needed in the proofs and allows to combine, in a straightforward way, different algorithmic ideas (e.g., adaptivity, optimism, implicit updates) and learning settings (e.g., strongly convex or composite objectives). This way we are able to reprove, extend and refine a large body of the literature, while keeping the proofs concise. The second contribution is a byproduct of this careful analysis: We present algorithms with improved variational bounds for smooth, composite objectives, including a new family of optimistic MD algorithms with only one projection step per round. Furthermore, we provide a simple extension of adaptive regret bounds to practically relevant non-convex problem settings with essentially no extra effort.

cs.LG↗

arXiv2017

Energy-Efficient Wireless Content Delivery with Proactive Caching

We propose an intelligent proactive content caching scheme to reduce the energy consumption in wireless downlink. We consider an online social network (OSN) setting where new contents are generated over time, and remain \textit{relevant} to the user for a random lifetime. Contents are downloaded to the user equipment (UE) through a time-varying wireless channel at an energy cost that depends on the channel state and the number of contents downloaded. The user accesses the OSN at random time instants, and consumes all the relevant contents. To reduce the energy consumption, we propose \textit{proactive caching} of contents under favorable channel conditions to a finite capacity cache memory. Assuming that the channel quality (or equivalently, the cost of downloading data) is memoryless over time slots, we show that the optimal caching policy, which may replace contents in the cache with shorter remaining lifetime with contents at the server that remain relevant longer, has certain threshold structure with respect to the channel quality. Since the optimal policy is computationally demanding in practice, we introduce a simplified caching scheme and optimize its parameters using policy search. We also present two lower bounds on the energy consumption. We demonstrate through numerical simulations that the proposed caching scheme significantly reduces the energy consumption compared to traditional reactive caching tools, and achieves close-to-optimal performance for a wide variety of system parameters.

cs.NI↗

arXiv2017

Following the Leader and Fast Rates in Linear Prediction: Curved Constraint Sets and Other Regularities

The follow the leader (FTL) algorithm, perhaps the simplest of all online learning algorithms, is known to perform well when the loss functions it is used on are convex and positively curved. In this paper we ask whether there are other "lucky" settings when FTL achieves sublinear, "small" regret. In particular, we study the fundamental problem of linear prediction over a non-empty convex, compact domain. Amongst other results, we prove that the curvature of the boundary of the domain can act as if the losses were curved: In this case, we prove that as long as the mean of the loss vectors have positive lengths bounded away from zero, FTL enjoys a logarithmic growth rate of regret, while, e.g., for polytope domains and stochastic data it enjoys finite expected regret. Building on a previously known meta-algorithm, we also get an algorithm that simultaneously enjoys the worst-case guarantees and the bound available for FTL.

cs.LG↗

arXiv2016

SDP Relaxation with Randomized Rounding for Energy Disaggregation

We develop a scalable, computationally efficient method for the task of energy disaggregation for home appliance monitoring. In this problem the goal is to estimate the energy consumption of each appliance over time based on the total energy-consumption signal of a household. The current state of the art is to model the problem as inference in factorial HMMs, and use quadratic programming to find an approximate solution to the resulting quadratic integer program. Here we take a more principled approach, better suited to integer programming problems, and find an approximate optimum by combining convex semidefinite relaxations randomized rounding, as well as a scalable ADMM method that exploits the special structure of the resulting semidefinite program. Simulation results both in synthetic and real-world datasets demonstrate the superiority of our method.

cs.LG↗

arXiv2016

Max-affine estimators for convex stochastic programming

In this paper, we consider two sequential decision making problems with a convexity structure, namely an energy storage optimization task and a multi-product assembly example. We formulate these problems in the stochastic programming framework and discuss an approximate dynamic programming technique for their solutions. As the cost-to-go functions are convex in these cases, we use max-affine estimates for their approximations. To train such a max-affine estimate, we provide a new convex regression algorithm, and evaluate it empirically for these planning scenarios.

math.OC↗

arXiv2016

Chaining Bounds for Empirical Risk Minimization

This paper extends the standard chaining technique to prove excess risk upper bounds for empirical risk minimization with random design settings even if the magnitude of the noise and the estimates is unbounded. The bound applies to many loss functions besides the squared loss, and scales only with the sub-Gaussian or subexponential parameters without further statistical assumptions such as the bounded kurtosis condition over the hypothesis class. A detailed analysis is provided for slope constrained and penalized linear least squares regression with a sub-Gaussian setting, which often proves tight sample complexity bounds up to logartihmic factors.

stat.ML↗

arXiv2015

Online Learning with Gaussian Payoffs and Side Observations

We consider a sequential learning problem with Gaussian payoffs and side information: after selecting an action $i$, the learner receives information about the payoff of every action $j$ in the form of Gaussian observations whose mean is the same as the mean payoff, but the variance depends on the pair $(i,j)$ (and may be infinite). The setup allows a more refined information transfer from one action to another than previous partial monitoring setups, including the recently introduced graph-structured feedback case. For the first time in the literature, we provide non-asymptotic problem-dependent lower bounds on the regret of any algorithm, which recover existing asymptotic problem-dependent lower bounds and finite-time minimax lower bounds available in the literature. We also provide algorithms that achieve the problem-dependent lower bound (up to some universal constant factor) or the minimax lower bounds (up to logarithmic factors).

stat.ML↗

arXiv2015

Fast Cross-Validation for Incremental Learning

Cross-validation (CV) is one of the main tools for performance estimation and parameter tuning in machine learning. The general recipe for computing CV estimate is to run a learning algorithm separately for each CV fold, a computationally expensive process. In this paper, we propose a new approach to reduce the computational burden of CV-based performance estimation. As opposed to all previous attempts, which are specific to a particular learning model or problem domain, we propose a general method applicable to a large class of incremental learning algorithms, which are uniquely fitted to big data problems. In particular, our method applies to a wide range of supervised and unsupervised learning tasks with different performance criteria, as long as the base learning algorithm is incremental. We show that the running time of the algorithm scales logarithmically, rather than linearly, in the number of CV folds. Furthermore, the algorithm has favorable properties for parallel and distributed implementation. Experiments with state-of-the-art incremental learning algorithms confirm the practicality of the proposed method.

stat.ML↗

arXiv2014

Adaptive Monte Carlo via Bandit Allocation

We consider the problem of sequentially choosing between a set of unbiased Monte Carlo estimators to minimize the mean-squared-error (MSE) of a final combined estimate. By reducing this task to a stochastic multi-armed bandit problem, we show that well developed allocation strategies can be used to achieve an MSE that approaches that of the best estimator chosen in retrospect. We then extend these developments to a scenario where alternative estimators have different, possibly stochastic costs. The outcome is a new set of adaptive Monte Carlo strategies that provide stronger guarantees than previous approaches while offering practical advantages.

cs.AI↗

arXiv2014

Efficient Multi-Start Strategies for Local Search Algorithms

Local search algorithms applied to optimization problems often suffer from getting trapped in a local optimum. The common solution for this deficiency is to restart the algorithm when no progress is observed. Alternatively, one can start multiple instances of a local search algorithm, and allocate computational resources (in particular, processing time) to the instances depending on their behavior. Hence, a multi-start strategy has to decide (dynamically) when to allocate additional resources to a particular instance and when to start new instances. In this paper we propose multi-start strategies motivated by works on multi-armed bandit problems and Lipschitz optimization with an unknown constant. The strategies continuously estimate the potential performance of each algorithm instance by supposing a convergence rate of the local search algorithm up to an unknown constant, and in every phase allocate resources to those instances that could converge to the optimum for a particular range of the constant. Asymptotic bounds are given on the performance of the strategies. In particular, we prove that at most a quadratic increase in the number of times the target function is evaluated is needed to achieve the performance of a local search algorithm started from the attraction region of the optimum. Experiments are provided using SPSA (Simultaneous Perturbation Stochastic Approximation) and k-means as local search algorithms, and the results indicate that the proposed strategies work well in practice, and, in all cases studied, need only logarithmically more evaluations of the target function as opposed to the theoretically suggested quadratic increase.

cs.LG↗

arXiv2013

Online Learning under Delayed Feedback

Online learning with delayed feedback has received increasing attention recently due to its several applications in distributed, web-based learning problems. In this paper we provide a systematic study of the topic, and analyze the effect of delay on the regret of online learning algorithms. Somewhat surprisingly, it turns out that delay increases the regret in a multiplicative way in adversarial problems, and in an additive way in stochastic problems. We give meta-algorithms that transform, in a black-box fashion, algorithms developed for the non-delayed case into ones that can handle the presence of delays in the feedback loop. Modifications of the well-known UCB algorithm are also developed for the bandit problem with delayed feedback, with the advantage over the meta-algorithms that they can be implemented with lower complexity.

cs.LG↗

arXiv2013

A Randomized Mirror Descent Algorithm for Large Scale Multiple Kernel Learning

We consider the problem of simultaneously learning to linearly combine a very large number of kernels and learn a good predictor based on the learnt kernel. When the number of kernels $d$ to be combined is very large, multiple kernel learning methods whose computational cost scales linearly in $d$ are intractable. We propose a randomized version of the mirror descent algorithm to overcome this issue, under the objective of minimizing the group $p$-norm penalized empirical risk. The key to achieve the required exponential speed-up is the computationally efficient construction of low-variance estimates of the gradient. We propose importance sampling based estimates, and find that the ideal distribution samples a coordinate with a probability proportional to the magnitude of the corresponding gradient. We show the surprising result that in the case of learning the coefficients of a polynomial kernel, the combinatorial structure of the base kernels to be combined allows the implementation of sampling from this distribution to run in $O(\log(d))$ time, making the total computational cost of the method to achieve an $ε$-optimal solution to be $O(\log(d)/ε^2)$, thereby allowing our method to operate for very large values of $d$. Experiments with simulated and real data confirm that the new algorithm is computationally more efficient than its state-of-the-art alternatives.

cs.LG↗

arXiv2012

Partition Tree Weighting

This paper introduces the Partition Tree Weighting technique, an efficient meta-algorithm for piecewise stationary sources. The technique works by performing Bayesian model averaging over a large class of possible partitions of the data into locally stationary segments. It uses a prior, closely related to the Context Tree Weighting technique of Willems, that is well suited to data compression applications. Our technique can be applied to any coding distribution at an additional time and space cost only logarithmic in the sequence length. We provide a competitive analysis of the redundancy of our method, and explore its application in a variety of settings. The order of the redundancy and the complexity of our algorithm matches those of the best competitors available in the literature, and the new algorithm exhibits a superior complexity-performance trade-off in our experiments.

cs.IT↗