Search

The advancement of autonomous capabilities in maritime navigation has gained significant attention, with a trajectory moving from decision support systems to full autonomy. This push towards autonomy has led to extensive research focusing on collision avoidance, a critical aspect of safe navigation. Among the various possible approaches, dynamic programming is a promising tool for optimizing collision avoidance maneuvers. This paper presents a DP formulation for the collision avoidance of autonomous vessels. We set up the problem framework, formulate it as a multi-stage decision process, define cost functions and constraints focusing on the actual requirements a marine maneuver must comply with, and propose a solution algorithm leveraging parallel computing. Additionally, we present a greedy approximation to reduce algorithm complexity. We put the proposed algorithms to the test in realistic navigation scenarios and also develop an extensive test on a large set of randomly generated scenarios, comparing them with the RRT* algorithm using performance metrics proposed in the literature. The results show the potential benefits of an autonomous navigation or decision support framework. [ABSTRACT FROM AUTHOR]

Many modern day cloud hosted applications such as virtual reality, real time games require low latency data access and computation to improve response time. So it is essential to bring the computation and data storage edge servers closer to the user's geographical location to improve response times and save bandwidth. In particulars, in online gaming and on demand video services, the required application data present at cloud servers need to be placed on the edge servers to provide low latency app-functionalities. The transfer of huge amount of data from cloud server to edge server incurs high cost and time penalties. Thus, we need an efficient way to solve edge data distribution (EDD) problem which distribute the application data to the edge servers that minimizes transfer cost. In this work, we provide a refined formulation of an optimal approach to solve the EDD problem using integer linear programming (ILP) technique. Due to the time complexity limitation of the ILP approach, we propose an O(k) approximation algorithm based on network Steiner tree estimation (EDD-NSTE) for estimating solutions to dense large-scale EDD problem. The proposed approach is analyzed to be 11/6 approximation which is better than the state-of-the-art 2 approximation EDD-A approach. The experimental evaluation through simulation using real world EUA data set demonstrate that the EDD-NSTE outperform state-of-the-art approach and other representative approaches. [ABSTRACT FROM AUTHOR]

We devise the first constant-factor approximation algorithm for finding an integral multi-commodity flow of maximum total value for instances where the supply graph together with the demand edges can be embedded on an orientable surface of bounded genus. This extends recent results for planar instances. Our techniques include an uncrossing algorithm, which is significantly more difficult than in the planar case, a partition of the cycles in the support of an LP solution into free homotopy classes, and a new rounding procedure for freely homotopic non-separating cycles. [ABSTRACT FROM AUTHOR]

The gate-based model is one of the leading quantum computing paradigms for representing quantum circuits. Within this paradigm, a quantum algorithm is expressed in terms of a set of quantum gates that are executed on the quantum hardware over time, subject to a number of constraints whose satisfaction must be guaranteed before running the circuit, to allow for feasible execution. The need to guarantee the previous feasibility condition gives rise to the Quantum Circuit Compilation Problem (QCCP). The QCCP has been demonstrated to be NP-Complete, and can be considered as a Planning and Scheduling problem. In this paper, we consider quantum compilation instances deriving from the general Quantum Approximation Optimization Algorithm (QAOA), applied to the MaxCut problem, devised to be executed on Noisy Intermediate Scale Quantum (NISQ) hardware architectures. More specifically, in addition to the basic QCCP version, we also tackle other variants of the same problem such as the QCCP-X (QCCP with crosstalk constraints), the QCCP-V (QCCP with variable qubit state initialization), as well as the QCCP-VX that includes both previous variants. All problem variants are solved using genetic algorithms. We perform an experimental study across a conventional set of instances taken from the literature, and show that the proposed genetic algorithm, termed G A VX , outperforms previous approaches in the literature. [ABSTRACT FROM AUTHOR]

The integrated industrial control systems require a reliable communication among each part of control systems. The limitation of network topology comprises the low latency of data transmission. The network algorithm topologies are sometimes failed to ensure the faulttolerant network design even if they offer it. Therefore, this works investigates a Node Insertion Algorithm (NIA) and Ant Colony Optimization (ACO) in order to solve the Travelling Salesman Problem (TSP) for network topologies in a communication network. The NP Non deterministic (polynomial time)-hardness based of TSP leads to the impossibility of polynomial time algorithm to solve an exponentially increasing amount of communication addresses. TSP itself is a well-known solution method known in the graph theory which is based on approximation algorithm. Therefore, this work also presents a comparison between the NIA and ACO for 200 nodes of network topologies. The results show that NIA is superior than ACO in terms of resulting ring's cost and computation duration. Three topologies are tested and it is observable that NIA requires only 14.88 seconds to finish the computation in which the resulting ring's length reaches 164019 units. Therefore, the achievement of the NIA can be applied for logical and physical nodes for a reliable connection in an industrial topology network. [ABSTRACT FROM AUTHOR]

How can we efficiently find very large numbers of clusters C in very large datasets N of potentially high dimensionality D? Here we address the question by using a novel variational approach to optimize Gaussian mixture models (GMMs) with diagonal covariance matrices. The variational method approximates expectation maximization (EM) by applying truncated posteriors as variational distributions and partial E-steps in combination with coresets. Run time complexity to optimize the clustering objective then reduces from O(NCD) per conventional EM iteration to O(N′G2D) for a variational EM iteration on coresets (with coreset size N ′ ≤ N and truncation parameter G ≪ C). Based on the strongly reduced run time complexity per iteration, which scales sublinearly with NC, we then provide a concrete, practically applicable, parallelized and highly efficient clustering algorithm. In numerical experiments on standard large-scale benchmarks we (A) show that also overall clustering times scale sublinearly with NC, and (B) observe substantial wall-clock speedups compared to already highly efficient recently reported results. The algorithm’s sublinear scaling allows for applications at scales where alternative methods cease to be applicable. We demonstrate such very large-scale applicability using the YFCC100M benchmark, for which we realize with a GMM of up to 50.000 clusters an optimization of a data density model with up to 150 M parameters. [ABSTRACT FROM AUTHOR]

Adversarial attacks have been extensively studied in recent years since they can identify the vulnerability of deep learning models before deployed. In this paper, we consider the black-box adversarial setting, where the adversary needs to craft adversarial examples without access to the gradients of a target model. Previous methods attempted to approximate the true gradient either by using the transfer gradient of a surrogate white-box model or based on the feedback of model queries. However, the existing methods inevitably suffer from low attack success rates or poor query efficiency since it is difficult to estimate the gradient in a high-dimensional input space with limited information. To address these problems and improve black-box attacks, we propose two prior-guided random gradient-free (PRGF) algorithms based on biased sampling and gradient averaging, respectively. Our methods can take the advantage of a transfer-based prior given by the gradient of a surrogate model and the query information simultaneously. Through theoretical analyses, the transfer-based prior is appropriately integrated with model queries by an optimal coefficient in each method. Extensive experiments demonstrate that, in comparison with the alternative state-of-the-arts, both of our methods require much fewer queries to attack black-box models with higher success rates. [ABSTRACT FROM AUTHOR]

How can we efficiently find very large numbers of clusters C in very large datasets N of potentially high dimensionality D? Here we address the question by using a novel variational approach to optimize Gaussian mixture models (GMMs) with diagonal covariance matrices. The variational method approximates expectation maximization (EM) by applying truncated posteriors as variational distributions and partial E-steps in combination with coresets. Run time complexity to optimize the clustering objective then reduces from O(NCD) per conventional EM iteration to O(N′G2D) for a variational EM iteration on coresets (with coreset size N ′ ≤ N and truncation parameter G ≪ C). Based on the strongly reduced run time complexity per iteration, which scales sublinearly with NC, we then provide a concrete, practically applicable, parallelized and highly efficient clustering algorithm. In numerical experiments on standard large-scale benchmarks we (A) show that also overall clustering times scale sublinearly with NC, and (B) observe substantial wall-clock speedups compared to already highly efficient recently reported results. The algorithm’s sublinear scaling allows for applications at scales where alternative methods cease to be applicable. We demonstrate such very large-scale applicability using the YFCC100M benchmark, for which we realize with a GMM of up to 50.000 clusters an optimization of a data density model with up to 150 M parameters. [ABSTRACT FROM AUTHOR]

Adversarial attacks have been extensively studied in recent years since they can identify the vulnerability of deep learning models before deployed. In this paper, we consider the black-box adversarial setting, where the adversary needs to craft adversarial examples without access to the gradients of a target model. Previous methods attempted to approximate the true gradient either by using the transfer gradient of a surrogate white-box model or based on the feedback of model queries. However, the existing methods inevitably suffer from low attack success rates or poor query efficiency since it is difficult to estimate the gradient in a high-dimensional input space with limited information. To address these problems and improve black-box attacks, we propose two prior-guided random gradient-free (PRGF) algorithms based on biased sampling and gradient averaging, respectively. Our methods can take the advantage of a transfer-based prior given by the gradient of a surrogate model and the query information simultaneously. Through theoretical analyses, the transfer-based prior is appropriately integrated with model queries by an optimal coefficient in each method. Extensive experiments demonstrate that, in comparison with the alternative state-of-the-arts, both of our methods require much fewer queries to attack black-box models with higher success rates. [ABSTRACT FROM AUTHOR]

This paper revisits the temporal difference (TD) learning algorithm for the policy evaluation tasks in reinforcement learning. Typically, the performance of TD(0) and TD($\lambda$ λ ) is very sensitive to the choice of stepsizes. Oftentimes, TD(0) suffers from slow convergence. Motivated by the tight link between the TD(0) learning algorithm and the stochastic gradient methods, we develop a provably convergent adaptive projected variant of the TD(0) learning algorithm with linear function approximation that we term AdaTD(0). In contrast to the TD(0), AdaTD(0) is robust or less sensitive to the choice of stepsizes. Analytically, we establish that to reach an $\epsilon$ ε accuracy, the number of iterations needed is $\tilde{O}(\epsilon ^{-2}\ln ^4\frac{1}{\epsilon }/\ln ^4\frac{1}{\rho })$ O ˜ (ε - 2 ln 4 1 ε / ln 4 1 ρ) in the general case, where $\rho$ ρ represents the speed of the underlying Markov chain converges to the stationary distribution. This implies that the iteration complexity of AdaTD(0) is no worse than that of TD(0) in the worst case. When the stochastic semi-gradients are sparse, we provide theoretical acceleration of AdaTD(0). Going beyond TD(0), we develop an adaptive variant of TD($\lambda$ λ ), which is referred to as AdaTD($\lambda$ λ ). Empirically, we evaluate the performance of AdaTD(0) and AdaTD($\lambda$ λ ) on several standard reinforcement learning tasks, which demonstrate the effectiveness of our new approaches. [ABSTRACT FROM AUTHOR]

Multi-relational learning on knowledge graphs infers high-order relations among the entities across the graphs. This learning task can be solved by label propagation on the tensor product of the knowledge graphs to learn the high-order relations as a tensor. In this paper, we generalize a widely used label propagation model to the normalized tensor product graph, and propose an optimization formulation and the scalable Low-rank Tensor-based Label Propagation algorithm (LowrankTLP) to infer multi-relations for two learning tasks, hyperlink prediction and multiple graph alignment. The optimization formulation minimizes the upper bound of the noisy-tensor estimating error for multiple graph alignment, by learning with a subset of the eigen-pairs in the spectrum of the normalized tensor product graph. We also provide a data-dependent transductive Rademacher bound for binary hyperlink prediction. We accelerate LowrankTLP with parallel tensor computation which enables label propagation on a tensor product of 100 graphs each of size 1000 in less than half hour in the simulation. LowrankTLP was also applied to predicting the author-paper-venue hyperlinks in publication records, alignment of segmented regions across up to 26 CT-scan images and alignment of protein-protein interaction networks across multiple species. The experiments demonstrate that LowrankTLP indeed well approximates the original label propagation with better scalability and accuracy. [ABSTRACT FROM AUTHOR]

One of the most extensively studied combinatorial optimization problems is the Travelling Salesman Problem (TSP). Considerable research efforts in the past have resulted in exact solvers. However, the runtime of such hand-crafted solutions increases exponentially with problem size. Ising model based solvers have also gained prominence due to their abilities to find fast and approximate solutions for combinatorial optimization problems. However, such Ising based heuristics also suffer from scalability as the solution quality becomes increasingly sub-optimal with increase in problem size. In this work, we propose Neuro-Ising – a machine learning framework which uses Ising models to find clusters of near-optimal partial solutions of large scale TSPs and combines those solutions by employing a supervised data driven mechanism, which we model as a Graph Neural Network (GNN). The GNN is trained from solution instances obtained through exact solvers and hence, the proposed approach generalizes to unseen problems while avoiding the run-time complexity otherwise required, if the solution is built from scratch. Using standard computing resources, our proposed framework rapidly converges to near-optimal solutions for 15 TSPs (upto $\sim 5k$ cities) from the TSPLib benchmark suite. We report $\sim 10.66\times $ speedup over Tabu Search for 8 problems. Furthermore, compared to two state-of-the-art clustering-based TSP solvers, Neuro-Ising achieves $\sim 38 \times $ faster convergence along with $\sim 8.9\%$ better quality of solution, on average. [ABSTRACT FROM AUTHOR]

Owing to its good compression ability, discrete cosine transform (DCT) is widely used in signal processing, including high-efficiency video coding (HEVC). For efficient implementation, approximate DCTs and integer DCTs were proposed, but the challenge in the tradeoff between coding performance and implementation cost always exists. To solve this problem, this article proposes a new algorithm to derive Int-DCT with good coding performance and low hardware cost. To address the multiobjective optimization problem, we extract information from the initially shortlisted transform matrices to derive the weighting factors for the different matrices properties. Those properties having a higher impact to the performance are assigned with higher weights and vice versa. Subsequently, an efficient search algorithm is proposed to shortlist those candidates which can lead to better coding performance evaluated by the proposed measure. Last but not the least, a splitting method is proposed to search the efficient extended double-base number system (EDBNS) representation of the coefficients for hardware implementation and select the solution which encounters the lowest cost. The experimental results in ASIC demonstrate that the implementation area and power cost of designs by the new algorithm are reduced by at least 11.1% and 17.5% for 32-point transform, respectively, over other competing algorithms. Meanwhile, the transforms generated by the proposed algorithm causes negligible impact to coding performance compared to the original transforms in HEVC. [ABSTRACT FROM AUTHOR]

In a large cloud data center HPC system, a critical problem is how to allocate the submitted tasks to heterogeneous servers that will achieve the goal of maximizing the system’s gain defined as the value of completed tasks minus system operation costs. We consider this problem in the online setting that tasks arrive in batches and propose a novel deep reinforcement learning (DRL) enhanced greedy optimization algorithm of two-stage scheduling interacting task sequencing and task allocation. For task sequencing, we deploy a DRL module to predict the best allocation sequence for each arriving batch of tasks based on the knowledge (allocation strategies) learnt from previous batches. For task allocation, we propose a greedy strategy that allocates tasks to servers one by one online following the allocation sequence to maximize the total gain increase. We show that our greedy strategy has a performance guarantee of competitive ratio $\frac{1}{1+\kappa }$ 1 1 + κ to the optimal offline solution, which improves the existing result for the same problem, where $\kappa$ κ is upper bounded by the maximum cost-to-gain ratio of each task. While our DRL module enhances the greedy algorithm by providing the likely-optimal allocation sequence for each batch of arriving tasks, our greedy strategy bounds DRL’s prediction error within a proven worst-case performance guarantee for any allocation sequence. It enables a better solution quality than that obtainable from both DRL and greedy optimization alone. Extensive experiment evaluation results in both simulation and real application environments demonstrate the effectiveness and efficiency of our proposed algorithm. Compared with the state-of-the-art baselines, our algorithm increases the system gain by about 10% to 30%. Our algorithm provides an interesting example of combining machine learning (ML) and greedy optimization techniques to improve ML-based solutions with a worst-case performance guarantee for solving hard optimization problems. [ABSTRACT FROM AUTHOR]

By employing time-varying proximal functions, adaptive subgradient methods (ADAGRAD) have improved the regret bound and been widely used in online learning and optimization. However, ADAGRAD with full matrix proximal functions (ADA-FULL) cannot handle large-scale problems due to the impractical $O(d^3)$ O (d 3) time and $O(d^2)$ O (d 2) space complexities, though it has better performance when gradients are correlated. In this paper, we propose two efficient variants of ADA-FULL via a matrix sketching technique called frequent directions (FD). The first variant named as ADA-FD directly utilizes FD to maintain and manipulate low-rank matrices, which reduces the space and time complexities to $O(\tau d)$ O (τ d) and $O(\tau ^2d)$ O (τ 2 d) respectively, where $d$ d is the dimensionality and $\tau \ll d$ τ ≪ d is the sketching size. The second variant named as ADA-FFD further adopts a doubling trick to accelerate FD used in ADA-FD, which reduces the average time complexity to $O(\tau d)$ O (τ d) while only doubles the space complexity of ADA-FD. Theoretical analysis reveals that the regret of ADA-FD and ADA-FFD is close to that of ADA-FULL as long as the outer product matrix of gradients is approximately low-rank. Experimental results demonstrate the efficiency and effectiveness of our algorithms. [ABSTRACT FROM AUTHOR]

This article investigates the finite-time distributed approximate optimization problem of higher order multiagent systems, where the local cost functions are considered to be quadratic functions. This problem is solved via penalty-function-based method. First, by the penalty-function method, a global approximate cost function is constructed. Second, nonlinear distributed optimization algorithms are proposed for higher order multiagent systems by the tool of adding a power integrator technique. In the optimization algorithms design, the gradients of the approximate cost function are utilized. Under the proposed optimization algorithms, the agents approach the approximate optimal solution in finite time. Although there exist errors (they may be called approximation errors) between the approximate minimizers and the global accurate minimizer, the approximation errors can be regulated by penalty parameter and the relationship between the bound of the approximation errors and the penalty parameter is given explicitly. Furthermore, the proposed distributed approximate optimization algorithms are applied to the optimal rendezvous problem of wheeled multimobile robots, making the mobile robots achieve approximate optimization rendezvous in finite time. The effectiveness of the proposed distributed optimization algorithms and their applications to optimal rendezvous problem are validated by simulations. [ABSTRACT FROM AUTHOR]

Fog networks facilitate ultra-low latency through the use of data availability near the network edge in fog servers. Existing work in fog networks considers the objective of energy efficiency and low latency for internet-of-things (IoT) for resource allocation. These works provide solutions to energy efficiency and low latency resource allocation problem without consideration of secure communication. This article investigates the benefits of fog architecture from the perspective of three promising technologies namely device-to-device (D2D) communication, caching, and physical layer security. We propose security provisioning followed by mode selection for D2D-assisted fog networks. The secrecy rate maximization problem is formulated first, which belongs to mixed-integer nonlinear programming (MINLP) problem. It is NP-hard, that is why an exhaustive search for finding the solution is complex. Keeping in view the complexity, a nonlinear technique namely outer approximation algorithm (OAA) is applied. OAA is a traditional algorithm, whose results are compared with the proposed heuristic algorithm, namely the security heuristic algorithm (SHA). Performance of the network is observed for the different numbers of eavesdroppers, IoT nodes, and fog nodes. [ABSTRACT FROM AUTHOR]

In this article, we discuss a sequential algorithm for the computation of a minimum-time speed profile over a given path, under velocity, acceleration, and jerk constraints. Such a problem arises in industrial contexts, such as automated warehouses, where LGVs need to perform assigned tasks as fast as possible in order to increase productivity. It can be reformulated as an optimization problem with a convex objective function, linear velocity and acceleration constraints, and nonconvex jerk constraints, which, thus, represent the main source of the difficulty. While existing nonlinear programming (NLP) solvers can be employed for the solution of this problem, it turns out that the performance and robustness of such solvers can be enhanced by the sequential line-search algorithm proposed in this article. At each iteration, a feasible direction, with respect to the current feasible solution, is computed, and a step along such direction is taken in order to compute the next iterate. The computation of the feasible direction is based on the solution of a linearized version of the problem, and the solution of the linearized problem, through an approach that strongly exploits its special structure, represents the main contribution of this work. The efficiency of the proposed approach with respect to existing NLP solvers is proven through different computational experiments. Note to Practitioners—This article was motivated by the needs of LGV manufacturers. In particular, it presents an algorithm for computing the minimum-time speed law for an LGV along a preassigned path, respecting assigned velocity, acceleration, and jerk constraints. The solution algorithm should be: 1) fast, since speed planning is made continuously throughout the workday, not only when an LGV receives a new task but also during the execution of the task itself, since conditions may change, e.g., if the LGV has to be halted for security reasons and 2) reliable, i.e., it should return solutions of high quality, because a better speed profile allows to save time and even small percentage improvements, say a 5% improvement, has a considerable impact on the productivity of the warehouse, and, thus, determines a significant economic gain. The algorithm that we propose meets these two requirements, and we believe that it can be a useful tool for LGV manufacturers and users. It is obvious that the proposed method also applies to the speed planning problem for vehicles other than LGVs, e.g., road vehicles. [ABSTRACT FROM AUTHOR]

With an increasing penetration of distributed energy resources (DERs), it has become a public interest regarding how to efficiently manage large-scale DERs into wholesale markets. The existing literature has focused on various coordination methods including iteration-based (e.g., Lagrangian relaxation) and projection-based (e.g., vertex enumeration) ones, which may not work for the cases with complicated constraints due to curse of dimensionality. In this article, we propose an efficient projection-based method that adapts to characterizing the feasible region of active distribution networks with temporal-coupled constraints. The problem of finding the temporal-coupled feasible region is formulated as a max–min optimization program, aiming to determine the largest set allowing zero-valued slack variables. Based on Duality Principle, the max–min model is transformed into a solvable mixed-integer linear programming that can be efficiently optimized by off-the-shelf solvers. In contrast to the existing methods searching for a vertex from interior points, we develop an outer progressive approximation algorithm that generates feasibility cuts for each iteration. Case studies based on a modified IEEE 33-bus feeder and Caracas 141-bus system demonstrate the effectiveness and efficiency of our proposed method in characterizing the feasible region with large-scale temporal-coupled constraints. [ABSTRACT FROM AUTHOR]

This paper explores the non-convex composition optimization consisting of inner and outer finite-sum functions with a large number of component functions. This problem arises in important applications such as nonlinear embedding and reinforcement learning. Although existing approaches such as stochastic gradient descent (SGD) and stochastic variance reduced gradient (SVRG) descent can be applied to solve this problem, their query complexities tend to be high, especially when the number of inner component functions is large. Therefore, to significantly improve the query complexity of current approaches, we have devised the stochastic composition via variance reduction (SCVR). What's more, we analyze the query complexity under different numbers of inner function and outer function. Based on different kinds of estimation of inner component function, we also present the SCVRII algorithm, though the order of query complexities are the same with SCVR. Additionally, we propose an extension to handle the mini-batch cases, which improve the query complexity under the optimal mini-batch size. The experimental results validate our proposed algorithms and theoretical analyses. [ABSTRACT FROM AUTHOR]

The Traveling Salesman Problem (TSP) aims to find the shortest tour for a salesman who starts and ends in the same city and visits the remaining n − 1 cities exactly once. There are a number of common generalizations of the problem including the Multiple Traveling Salesman Problem (MTSP), where instead of one salesman, there are k salesmen and the same amount of individual tours are to be constructed. We consider the Euclidean version of the problem where the distances between the cities are calculated in two-dimensional Euclidean space. Both general the TSP and its Euclidean version are strongly NP-hard. Hence, approximation algorithms with a good practical behavior are of primary interest. We describe a general method for the solution of the Euclidean versions of the TSP (including MTSP) that yields approximation algorithms with a favorable practical behavior for large real-life instances. Our method creates special types of convex hulls, which serve as a basis for the constructions of our initial and intermediate partial solutions. Here, we overview three algorithms; one of them is for the bounded version of the MTSP. The proposed novel algorithm for the Euclidean TSP provides close-to-optimal solutions for some real-life instances. [ABSTRACT FROM AUTHOR]

Reorganizing bus frequencies to cater for the actual travel demands can significantly save the cost of the public transport system. Many, if not all, previous studies formulate this as a bus frequency optimization problem that tries to minimize passengers’ average waiting time. On the other hand, many investigations have confirmed that the user satisfaction drops faster as the waiting time increases. Consequently, this paper studies the bus frequency optimization problem considering the user satisfaction. Specifically, for the first time to our best knowledge, we study how to schedule the buses such that the total number of passengers who could receive their bus services within the waiting time threshold can be maximized. We propose two variants of the problem, FAST and FASTCO, to cater for different application needs and prove that both are NP-hard. To solve FAST effectively and efficiently, we first present an index-based $(1-1/e)$ (1 - 1 / e) -approximation algorithm. By exploiting the locality property of routes in a bus network, we further propose a partition-based greedy method that achieves a $(1-\rho)(1-1/e)$ (1 - ρ) (1 - 1 / e) approximation ratio. Then we propose a progressive partition-based greedy method to further boost the efficiency while achieving a $(1-\rho)(1-1/e-\varepsilon)$ (1 - ρ) (1 - 1 / e - ɛ) approximation ratio. For the FASTCO problem, two greedy-based heuristic methods are proposed. Experiments on a real city-wide bus dataset in Singapore have been conducted to verify the efficiency, effectiveness, and scalability of our methods in addressing FAST and FASTCO respectively. [ABSTRACT FROM AUTHOR]

Revenue maximization (RM) is one of the most important problems in social networks, which attempts to find a small subset of users that make the expected revenue maximized. It has been studied in depth before. However, most of the existing literature was based on nonadaptive seeding strategies and simple information diffusion models. It considered the number of influenced users as a measurement unit to quantify the revenue. Until the emergence of the collaborate game model, it considered the activity as a basic object to compute the revenue. An activity initiated by a user can only influence those users whose distances are within ${k}$ -hop from the initiator. Based on that, we adopt an adaptive seed strategy and formulate an RM under the size budget (RMSB) problem. If taking into account the product’s promotion, we extend it to an RM under the community budget problem, where the influence can be distributed over the whole network uniformly. We can prove that our objective function is adaptive monotone and not adaptive submodular, but it is adaptive submodular in some special cases. We study these two problems under both the special submodular cases and general nonsubmodular cases, and propose RMSBSolver and RMCBSolver to solve them with strong theoretical guarantees, respectively. In particular, we give a data-dependent approximation ratio by adaptive primal curvature for the RMSB in general nonsubmodular cases. Finally, we evaluate our proposed algorithms by conducting experiments on real datasets, and show the effectiveness and accuracy of our solutions. [ABSTRACT FROM AUTHOR]

We introduce a novel domination problem, which we call Dual Domination (DD). We assume that the nodes in a network are partitioned into two categories: Positive nodes (V +) and negative nodes (V −). We study the Maximum Bounded Dual Domination , where, given a bound k , the problem is to find a set D ⊆ V + , which maximizes the number of nodes dominated in V + , dominating at most k nodes in V −. We show that such problem is hard to approximate to a factor better than (1 − 1 / e). We give a polynomial time algorithm with approximation guaranteed (1 − e − 1 / Δ) , where Δ represents the maximum number of neighbors in V + of any node in V − , and an O (| V | k 2) time algorithm to solve the problem on trees. We also study two related problems named Maximum Dual Domination and minimum Negative Dual Domination. For both problems, we show hardness results and provide O (| V |) time algorithms on trees. [ABSTRACT FROM AUTHOR]

Recent advances in quantum technologies pave the way for noisy intermediate-scale quantum (NISQ) devices, where the quantum approximation optimization algorithm (QAOA) constitutes a promising candidate for demonstrating tangible quantum advantages based on NISQ devices. In this paper, we consider the maximum likelihood (ML) detection problem of binary symbols transmitted over a multiple-input and multiple-output (MIMO) channel, where finding the optimal solution is exponentially hard using classical computers. Here, we apply the QAOA for the ML detection by encoding the problem of interest into a level- $p$ QAOA circuit having $2p$ variational parameters, which can be optimized by classical optimizers. This level- $p$ QAOA circuit is constructed by applying the prepared Hamiltonian to our problem and the initial Hamiltonian alternately in $p$ consecutive rounds. More explicitly, we first encode the optimal solution of the ML detection problem into the ground state of a problem Hamiltonian. Using the quantum adiabatic evolution technique, we provide both analytical and numerical results for characterizing the evolution of the eigenvalues of the quantum system used for ML detection. Then, for level-1 QAOA circuits, we derive the analytical expressions of the expectation values of the QAOA and discuss the complexity of the QAOA based ML detector. Explicitly, we evaluate the computational complexity of the classical optimizer used and the storage requirement of simulating the QAOA. Finally, we evaluate the bit error rate (BER) of the QAOA based ML detector and compare it both to the classical ML detector and to the classical minimum mean squared error (MMSE) detector, demonstrating that the QAOA based ML detector is capable of approaching the performance of the classical ML detector. [ABSTRACT FROM AUTHOR]

Toward better approximation for the minimum-weighted vertex cover (MWVC) problem in multiagent systems, we present a distributed algorithm from the perspective of learning in games. For self-organized coordination and optimization, we see each vertex as a potential game player who makes decisions using local information of its own and the immediate neighbors. The resulting Nash equilibrium is classified into two categories, i.e., the inferior Nash equilibrium (INE) and the dominant Nash equilibrium (DNE). We show that the optimal solution must be a DNE. To achieve better approximation ratios, local rules of perturbation and weighted memory are designed, with the former destroying the stability of an INE and the latter facilitating the refinement of a DNE. By showing the existence of an improvement path from any INE to a DNE, we prove that when the memory length is larger than 1, our algorithm converges in finite time to DNEs, which could not be improved by exchanging the action of a selected node with all its unselected neighbors. Moreover, additional freedom for solution efficiency refinement is provided by increasing the memory length. Finally, intensive comparison experiments demonstrate the superiority of the presented methodology to the state of the art, both in solution efficiency and computation speed. [ABSTRACT FROM AUTHOR]

We consider decentralized large-scale continuous submodular constrained optimization problems over networks, where the goal is to maximize a sum of nonconvex functions with diminishing returns property. However, the computations of the projection step and the whole gradient can become prohibitive in high-dimensional constrained optimization problems. For this reason, a decentralized randomized block-coordinate Frank-Wolfe algorithm is proposed for submoduar maximization over networks by local communication and computation, which adopts the randomized block-coordinate descent and the Frank-Wolfe technique. We also show that the proposed algorithm converges to an approximation fact $(1-e^{-p_{\max }/p_{\min }})$ of the global maximal points at a rate of $\mathcal {O}(1/T)$ by choosing a suitable stepsize, where $T$ is the number of iterations. In addition, we confirm the theoretical results by experiments. [ABSTRACT FROM AUTHOR]

The manta ray foraging optimization algorithm (MRFO) is one of the promised meta-heuristic optimization algorithms. However, it can stick to a local minimum, consuming iterations without reaching the optimum solution. So, this paper proposes a hybridization between MRFO, and the genetic algorithm (GA) based on a pseudo parameter; where the GA can help MRFO to escape from falling into the local minimum. It is called a pseudo genetic algorithm with manta-ray foraging optimization (PGA-MRFO). The proposed algorithm is not a classical hybridization between MRFO and GA, wherein the classical hybridization consumes time in the search process as each algorithm is applied to all system variables. In addition, the classical hybridization results in an extended search algorithm, especially in systems with many variables. The PGA-MRFO hybridizes the pseudo-parameter-based GA and the MRFO algorithm to produce a more efficient algorithm that combines the advantages of both algorithms without getting stuck in a local minimum or taking a long time in the calculations. The pseudo parameter enables the GA to be applied to a specific number of variables and not to all system variables leading to reduce the computation time and burden. Also, the proposed algorithm used an approximation for the gradient of the objective function, which leads to dispensing derivatives calculations. Besides, PGA-MRFO depends on the pseudo inverse of non-square matrices, which saves calculations time; where the dependence on the pseudo inverse gives the algorithm more flexibility to deal with square and non-square systems. The proposed algorithm will be tested on the test functions that the main MRFO failed to find their optimum solution to prove its capability and efficiency. In addition, it will be applied to solve the unit commitment (UC) problem as one of the vital power system problems to show the validity of the proposed algorithm in practical applications. Finally, several analyses will be applied to the proposed algorithm to illustrate its effectiveness and reliability. [ABSTRACT FROM AUTHOR]

Most existing random access schemes for machine-type communications (MTC) simply adopt a uniform preamble selection distribution, irrespective of the underlying device activity distributions. Hence, they may yield unsatisfactory access efficiency. In this paper, we model device activities for MTC as multiple Bernoulli random variables following an arbitrary multivariate Bernoulli distribution which can reflect both dependent and independent device activities. Then, we optimize preamble selection and access barring for random access in MTC according to the underlying joint device activity distribution. Specifically, we investigate three cases of the joint device activity distribution, i.e., the cases of perfect, imperfect, and unknown joint device activity distributions, and formulate the average, worst-case average, and sample average throughput maximization problems, respectively. The problems in the three cases are challenging nonconvex problems. In the case of perfect joint device activity distribution, we develop an iterative algorithm and a low-complexity iterative algorithm to obtain stationary points of the original problem and an approximate problem, respectively. In the case of imperfect joint device activity distribution, we develop an iterative algorithm and a low-complexity iterative algorithm to obtain a Karush-Kuhn-Tucker (KKT) point of an equivalent problem and a stationary point of an approximate problem, respectively. Finally, in the case of unknown joint device activity distribution, we develop an iterative algorithm to obtain a stationary point. The proposed solutions are widely applicable and outperform existing solutions for dependent and independent device activities. [ABSTRACT FROM AUTHOR]

Hammerstein model with a static nonlinearity followed by a linear filter is commonly used in numerous applications. This paper focuses on adaptive filtering techniques for parameter identification of Hammerstein systems and output prediction of nonlinear systems. By formulating the underlying filtering problem as a recursive bilinear least-squares optimization with the non-convex feasible region constraint, we develop a recursive non-convex projected least-squares (RncPLS) algorithm based on alternating direction method of multipliers (ADMM). The RncPLS algorithm alternates between implementing ridge regression and projecting on the non-convex feasible set, which successively refines the system parameters. The convergence and accuracy properties of the proposed RncPLS algorithm are theoretically investigated. Moreover, extensive simulation results in the context of system identification, nonlinear predication, and acoustic echo cancellation, are also included to demonstrate the performance characteristics of the proposed algorithm. [ABSTRACT FROM AUTHOR]

Numerical optimization has been playing an increasingly important role in the design of contemporary antenna systems. Due to the shortage of design-ready theoretical models, optimization is mainly based on electromagnetic (EM) analysis, which tends to be costly. Numerous techniques have evolved to abate this cost, including surrogate-assisted frameworks for global optimization, or sparse sensitivity updates for speeding up local search. In the latter, CPU-heavy updates of the system response sensitivity through finite differentiation are suppressed based on, e.g., the magnitude of design variability during the optimization run. Another approach is to incorporate variable-resolution simulations. Recently, a technique exploiting a continuous spectrum of admissible model fidelity levels has been reported, thereby allowing for a considerable reduction of computational expenditures. Seeking further savings, this work introduces an accelerated gradient-based algorithm with sparse sensitivity updates and variable-resolution EM simulations. Our technique is validated using four broadband antennas, and demonstrated to offer substantial (around 80%) savings over the benchmark while maintaining acceptable design quality. [ABSTRACT FROM AUTHOR]

Generation Expansion Planning (GEP) can inform regulation, electricity market design, and regional system planning by identifying adaptive investment strategies. Relevant uncertainties include hourly variability in load and renewable generation and decadal-scale uncertainty in technology, markets, and regulation. A multi-stage and multi-scale stochastic GEP model that represents these uncertainties at sufficient resolution becomes intractable. We present an approach for representing this multi-scale uncertainty, and compare it to existing methods, applied to a two-stage stochastic GEP model with a cumulative carbon emission target. For long-term uncertainty, we compare partitioning methods, which reduce the number of decision variables but retain all scenarios, to representative scenario methods, which retain only a subset of the original scenarios. For short-term uncertainty, we compare methods that select representative weeks based on distance metrics in the parameter space to methods that use the covariance of outcomes across feasible decisions to select weeks. We find that scenario reduction methods struggle to find the appropriate investment levels for variable renewable generation and consequently produce more costly plans than scenario partitioning methods. While simple approximating methods perform well with larger models, covariance-based approximations have the best performance overall. [ABSTRACT FROM AUTHOR]

Gas-fired generators, with their ability to quickly ramp up and down their electricity production, play an important role in managing renewable energy variability. However, these changes in electricity production translate into variability in the consumption of natural gas, and propagation of uncertainty from the electric grid to the natural gas system. To ensure that both systems are operating safely, there is an increasing need for coordination and uncertainty management among the electricity and gas networks. A challenging aspect of this coordination is the consideration of natural gas dynamics, which play an important role in intra-day operation, but give rise to a set of non-linear and non-convex equations that are hard to optimize even in the deterministic case. Ideally, the problem is formulated as a stochastic problem but many conventional methods for stochastic optimization are numerically intractable because they either incorporate a large number of scenarios directly or require the underlying problem to be convex. To address these challenges, we propose using a Stochastic Hybrid Approximation algorithm to more efficiently solve these problems and investigate the efficacy of several different variants of this algorithm. Our case study demonstrates that the proposed technique is able to quickly obtain high quality solutions and outperforms existing benchmarks such as Generalized Benders Decomposition. We demonstrate that coordinated uncertainty management that accounts for the gas system can significantly reduce both electric and gas system load shed in stressed conditions. [ABSTRACT FROM AUTHOR]

Graph matching, the problem of aligning a pair of graphs so as to minimize their edge disagreements, has received widespread attention owing to its broad spectrum of applications in data science. As the problem is NP–hard in the worst-case, a variety of approximation algorithms have been proposed for obtaining high quality, suboptimal solutions. In this article, we approach the task of designing an efficient polynomial-time approximation algorithm for graph matching from a previously unconsidered perspective. Our key result is that graph matching can be formulated as maximizing a monotone, supermodular set function subject to matroid intersection constraints. We leverage this fact to apply a discrete optimization variant of the minorization-maximization algorithm which exploits supermodularity of the objective function to iteratively construct and maximize a sequence of global lower bounds on the objective. At each step, we solve a maximum weight matching problem in a bipartite graph. Differing from prior approaches, the algorithm exploits the combinatorial structure inherent in the problem to generate a sequence of iterates featuring monotonically non-decreasing objective value while always adhering to the combinatorial matching constraints. Experiments on real-world data demonstrate the empirical effectiveness of the algorithm relative to the prevailing state-of-the-art. [ABSTRACT FROM AUTHOR]

Despite being commonly used in big-data analytics; the outcome of dimensionality reduction remains a black-box to most of its users. Understanding the quality of a low-dimensional embedding is important as not only it enables trust in the transformed data, but it can also help to select the most appropriate dimensionality reduction algorithm in a given scenario. As existing research primarily focuses on the visual exploration of embeddings, there is still a need for enhancing interpretability of such algorithms. To bridge this gap, we propose two novel interactive explanation techniques for low-dimensional embeddings obtained from any dimensionality reduction algorithm. The first technique LAPS produces a local approximation of the neighborhood structure to generate interpretable explanations on the preserved locality for a single instance. The second method GAPS explains the retained global structure of a high-dimensional dataset in its embedding, by combining non-redundant local-approximations from a coarse discretization of the projection space. We demonstrate the applicability of the proposed techniques using 16 real-life tabular, text, image, and audio datasets. Our extensive experimental evaluation shows the utility of the proposed techniques in interpreting the quality of low-dimensional embeddings, as well as with selecting the most suitable dimensionality reduction algorithm for any given dataset. [ABSTRACT FROM AUTHOR]

Existing wideband beampattern synthesis methods usually require multiple beamformers, each corresponding to one frequency bin, which results in excessive resource consumption in engineering practice. To alleviate the resource consumption of wideband array beamforming, we propose a novel time-domain wideband digital beamformer that can control the beam shape over the whole bandwidth for interference mitigation by using only one set of amplitude–phase weights and integer time delay filters. To tackle the resultant optimization problem, two effective algorithms are introduced, and both can solve the problem within polynomial time. One is an alternating optimization algorithm that decomposes the original problem into multiple tractable subproblems iteratively. The other is the convex approximation algorithm that transforms the original problem into a series of second-order cone programming problems iteratively. The two algorithms exhibit their respective advantages in terms of convergence and computational complexity. Extensive numerical examples are presented to evaluate the proposed time-domain wideband beamforming strategy, highlighting that it can form a deep notch in any interested spatial–spectral region to suppress the strong interferences. [ABSTRACT FROM AUTHOR]

Mobile Edge Computing (MEC) has become an attractive solution to enhance the computing and storage capacity of mobile devices by leveraging available resources on edge nodes. In MEC, the arrivals of tasks are highly dynamic and are hard to predict precisely. It is of great importance yet very challenging to assign the tasks to edge nodes with guaranteed system performance. In this article, we aim to optimize the revenue earned by each edge node by optimally offloading tasks to the edge nodes. We formulate the revenue-driven online task offloading (ROTO) problem, which is proved to be NP-hard. We first relax ROTO to a linear fractional programming problem, for which we propose the Level Balanced Allocation (LBA) algorithm. We then show the performance guarantee of LBA through rigorous theoretical analysis, and present the LB-Rounding algorithm for ROTO using the primal-dual technique. The algorithm achieves an approximation ratio of $2(1+\xi)\ln (d+1)$ 2 (1 + ξ) ln (d + 1) with a considerable probability, where $d$ d is the maximum number of process slots of an edge node and $\xi$ ξ is a small constant. The performance of the proposed algorithm is validated through both trace-driven simulations and testbed experiments. Results show that our proposed scheme is more efficient compared to baseline algorithms. [ABSTRACT FROM AUTHOR]

Dual Coordinate Descent (DCD) is one of the most popular optimization methods. The parallelization of DCD is difficult, as DCD is sequential in nature. As such, simultaneously running multiple DCD threads on batches of data elements causes result inaccuracy and slow convergence, due to the concurrent updates of multiple coordinates. Some parallelization methods adopt separable approximate functions that depend on the degree of parallelism. Such dependencies result in both poor scalability and slow-convergence. To address these challenges, in this paper we present a new parallel primal-dual algorithm for DCD, called PPD. In PPD, the block data distribution is utilized to obtain a new approximate function that is independent from the parallelism. Moreover, PPD is designed with a novel primal-dual acceleration scheme to approach the optimal solution closely and quickly. We demonstrate the advantages of PPD in terms of scalability and efficiency through experiments. [ABSTRACT FROM AUTHOR]

Area coverage and network lifetime are two contradictory issues to the architecture development of a wireless sensor network (WSN). A satisfactory balance could be achieved by deploying abundant sensor nodes randomly and dividing them into $k$ exclusive cover sets. Toward self-organized partition with higher efficiency, we address the problem from the perspective of networked potential games and propose a memorial mixed-response algorithm (MMRA), which is implemented in a distributed and synchronous manner. Being viewed as a game player, each sensor node first updates its memory using a temporary action, which is generated by following a mixed response rule. After this, the coordination evolves into the next iteration by each player randomly drawing an action from its memory with equal probabilities. We prove that our algorithm converges with probability 1 to a convention of Nash equilibria, with the worst approximation ratio strictly larger than 0.5. Moreover, it is also found that a tradeoff between solution efficiency and computation time could be achieved via the adjustment of the amount of randomness introduced via the memory length $m$ as well as the probability $p_{m}$ , where better partition results are more likely to be generated using a larger $m$ and smaller $p_{m}$. Comparisons with existing distributed methods demonstrate the superiority of our method in terms of solution refinement as well as convergence speed. [ABSTRACT FROM AUTHOR]

Non-orthogonal multiple access (NOMA) and backscatter communication are two emerging technologies for low-power communication. In this paper, we consider a NOMA backscatter system, where signals from two backscatter devices are multiplexed on a frequency resource block using NOMA in each time slot. Our objective is to maximize the average energy efficiency by optimizing backscatter device pairing, reflection coefficients of backscatter devices, and the transmit power of the reader. We formulate the average energy efficiency maximization problem subject to the minimum circuit power and the minimum data rate requirements of the backscatter devices, and the transmit power constraint of the reader. The formulated problem is nonconvex. To obtain a suboptimal solution for this problem, we use alternating optimization technique and decompose the problem into two subproblems. The subproblems are solved by using fractional programming, Dinkelbach’s algorithm, and successive convex approximation method. Simulation results show that our proposed algorithm converges quickly to a suboptimal solution. Our proposed algorithm outperforms several baseline algorithms, including the genetic algorithm, fixed device pairing scheme, conventional device pairing scheme, maximum transmit power allocation scheme, and random reflection coefficient selection scheme, in terms of the average energy efficiency. The optimality gap of our proposed algorithm is investigated by comparing with the optimal scheme, in which the optimal device pairing is obtained based on exhaustive search. [ABSTRACT FROM AUTHOR]

Data representation techniques have made a substantial contribution to advancing data processing and machine learning (ML). Improving predictive power was the focus of previous representation techniques, which unfortunately perform rather poorly on the interpretability in terms of extracting underlying insights of the data. Recently, the Kolmogorov model (KM) was studied, which is an interpretable and predictable representation approach to learning the underlying probabilistic structure of a set of random variables. The existing KM learning algorithms using semi-definite relaxation with randomization (SDRwR) or discrete monotonic optimization (DMO) have, however, limited utility to big data applications because they do not scale well computationally.In this paper, we propose a computationally scalable KM learning algorithm, based on the regularized dual optimization combined with enhanced gradient descent (GD) method. To make our method more scalable to large-dimensional problems, we propose two acceleration schemes, namely, the eigenvalue decomposition (EVD) elimination strategy and an approximate EVD algorithm. Furthermore, a thresholding technique by exploiting the error bound analysis and leveraging the normalized Minkowski $\ell _1$ -norm, is provided for the selection of the number of iterations of the approximate EVD algorithm. When applied to big data applications, it is demonstrated that the proposed method can achieve compatible training/prediction performance with significantly reduced computational complexity; roughly two orders of magnitude improvement in terms of the time overhead, compared to the existing KM learning algorithms. Furthermore, it is shown that the accuracy of logical relation mining for interpretability by using the proposed KM learning algorithm exceeds 80%. [ABSTRACT FROM AUTHOR]

Addressing non-convexity plays a fundamental role in solving the optimal electricity-gas flow models. In this paper, an improved spatial branch-and-bound algorithm is proposed to solve the non-convex problem, which is formulated as a mixed-integer bilinear programming, for its exact solution. The core of the algorithm is to divide the non-convex model into convex and small sub-models by branching on specific continuous variables, so that the non-convex problem can be equivalent to a rooted tree for exploration. The exactness of the algorithm is guaranteed by the same criterion as the classical branch-and-bound algorithm. To alleviate the computational burden, a novel two-stage spatial branching strategy is developed to improve the effectiveness and efficiency of the branching operations. The performance of the proposed algorithm is verified on two integrated electricity-gas systems with different sizes. Numerical results demonstrate that our method achieves a balance among feasibility, optimality, and efficiency. The comparison with another 6 convexification-based methods, 3 state-of-the-art non-convex optimization solvers, and 2 spatial branch-and-bound algorithms with classical branching rules further shows the superiority of our algorithm. [ABSTRACT FROM AUTHOR]

Learning to improve AUC performance for imbalanced data is an important machine learning research problem. Most methods of AUC maximization assume that the model function is linear in the original feature space. However, this assumption is not suitable for nonlinear separable problems. Although there have been some nonlinear methods of AUC maximization, scaling up nonlinear AUC maximization is still an open question. To address this challenging problem, in this paper, we propose a novel large-scale nonlinear AUC maximization method (named as TSAM) based on the triply stochastic gradient descents. Specifically, we first use the random Fourier feature to approximate the kernel function. After that, we use the triply stochastic gradients w.r.t. the pairwise loss and random feature to iteratively update the solution. Finally, we prove that TSAM converges to the optimal solution with the rate of $ \mathcal {O}(1/t)$ O (1 / t) after $t$ t iterations. Experimental results on a variety of benchmark datasets not only confirm the scalability of TSAM, but also show a significant reduction of computational time compared with existing batch learning algorithms, while retaining the similar generalization performance. [ABSTRACT FROM AUTHOR]

We consider stochastic algorithms derived from methods for solving deterministic optimization problems, especially comparison-based algorithms derived from stochastic approximation algorithms with a constant step-size. We develop a methodology for proving geometric convergence of the parameter sequence { θ n } n ⩾ 0 of such algorithms. We employ the ordinary differential equation (ODE) method, which relates a stochastic algorithm to its mean ODE, along with a Lyapunov-like function Ψ such that the geometric convergence of Ψ (θ n) implies – in the case of an optimization algorithm – the geometric convergence of the expected distance between the optimum and the search point generated by the algorithm. We provide two sufficient conditions for Ψ (θ n) to decrease at a geometric rate: Ψ should decrease "exponentially" along the solution to the mean ODE, and the deviation between the stochastic algorithm and the ODE solution (measured by Ψ) should be bounded by Ψ (θ n) times a constant. We also provide practical conditions under which the two sufficient conditions may be verified easily without knowing the solution of the mean ODE. Our results are any-time bounds on Ψ (θ n) , so we can deduce not only the asymptotic upper bound on the convergence rate, but also the first hitting time of the algorithm. The main results are applied to a comparison-based stochastic algorithm with a constant step-size for optimization on continuous domains. [ABSTRACT FROM AUTHOR]

There is a large growth in hardware and software systems capable of producing vast amounts of image and video data. These systems are rich sources of continuous image and video streams. This motivates researchers to build scalable computer vision systems that utilize data-streaming concepts for processing of visual data streams. However, several challenges exist in building large-scale computer vision systems. For example, computer vision algorithms have different accuracy and speed profiles depending on the content, type, and speed of incoming data. Also, it is not clear how to adaptively tune these algorithms in large-scale systems. These challenges exist because we lack formal frameworks for building and optimizing large-scale visual processing. This paper presents formal methods and algorithms that aim to overcome these challenges and improve building and optimizing large-scale computer vision systems. We describe a formal algebra framework for the mathematical description of computer vision pipelines for processing image and video streams. The algebra naturally describes feedback control and provides a formal and abstract method for optimizing computer vision pipelines. We then show that a general optimizer can be used with the feedback-control mechanisms of our stream algebra to provide a common online parameter optimization method for computer vision pipelines. [ABSTRACT FROM AUTHOR]

Constrained reinforcement learning (CRL), also termed as safe reinforcement learning, is a promising technique enabling the deployment of RL agent in real-world systems. In this paper, we propose a successive convex approximation based off-policy optimization (SCAOPO) algorithm to solve the general CRL problem, which is formulated as a constrained Markov decision process (CMDP) in context of the average cost. The SCAOPO is based on solving a sequence of convex objective/feasibility optimization problems obtained by replacing the objective and constraint functions in the original problem with convex surrogate functions. The proposed SCAOPO enables reuse of experiences from previous updates, thereby significantly reducing implementation cost when deployed in real-world engineering systems that need to online learn the environment. In spite of the time-varying state distribution and the stochastic bias incurred by off-policy learning, the SCAOPO with a feasible initial point can still provably converge to a Karush-Kuhn-Tucker (KKT) point of the original problem almost surely. [ABSTRACT FROM AUTHOR]