Optimal & near-optimal space efficient decision diagram algorithms

This special issue of the VLSI DESIGN: an International Journal of Custom-Chip Design, Simulation and Testing is devoted to spectral techniques, new types of decision trees and diagrams, their applications, efficient ways of their calculation and mutual relations between spectral techniques and classical decision diagrams. Manipulations and calculations with discrete functions are fundamental tasks in many areas of Computer Science and Engineering. Many problems in digital system design, simulation and testing can be expressed as a sequence of operations on discrete functions. The performance of Computer-Aided Design systems used in solving various problems in this area strongly depends on the efficiency of representations of discrete functions. Decision Diagrams (DDs) have proved very convenient data structure for discrete function representations, permitting manipulations and calculations with large functions efficiently in terms of time and space. In many applications, as for example those involving large matrices, conventional algorithms are significantly improved by using DDs. In logic design, such applications relate to basic problems of the design, verification, simulation and testing logical networks. Besides Boolean functions, some other types of discrete representations, such as multi-valued functions, cube sets, EXOR-based spectral and probabilistic formulas have to be dealt with. Hence the numbers of different decision trees and diagrams have been introduced for these various applications and representations. The spectral or polynomial representation of Boolean function is obtained by expressing it in terms of an orthogonal basis. The most popular basis is formed by discrete Walsh functions. Note that the spectral and the polynomial representation of a Boolean function are equivalent. The discrete orthogonal functions are referred to as the basis functions when the spectral or abstract harmonic terminology is used and as monomials when the polynomial terminology is used. Such polynomial representations have extensive applications in coding theory and cryptography as well as design of nonlinear filters. Spectral techniques based on Walsh, Haar, Arithmetic and Reed–Muller transforms have been used in digital logic design for more than 30 years. These techniques have been used for Boolean function classification, disjoint decomposition, identification of symmetries, parallel and serial linear decomposition, spectral translation synthesis (extraction of linear preand post-filters), multiplexer synthesis, threshold logic synthesis, state assignment and testing, and evaluation of logic complexity. The renewed interest in applications of spectral methods in design of VLSI digital circuits is caused by their excellent design for testability properties and in their efficiency in Boolean mapping problems. Furthermore, the practical applicability of these techniques was greatly improved with the development of efficient methods that allow calculating and operating on spectra of Boolean functions directly from reduced representations of such functions in the form of arrays of cubes or decision diagrams. Following from these recent developments, techniques have been presented for efficient spectral translation to decompose a circuit into a cascade of two sub circuits: linear block composed of EXOR gates fed by the primary input of the overall circuit. Such linear decomposition drastically simplifies the synthesis task. The choice of suitable linear transformation is based on a complexity measure assigned to each Boolean function that is heuristically related to the complexity of the final circuit implementation. To avoid the experimental costs of computing the Walsh transform, Binary Decision Diagrams (BDDs) based techniques are used. Recently the idea of linear transformations is applied to decision diagrams. By combining powerful spectral linear techniques with variable reordering techniques, it is possible to synthesize large Boolean functions with standard Computer-Aided Design tools that fail otherwise. The areas addressed by this special issue span spectral transforms, their applications and efficient calculation methods through decision diagrams. New spectral decision diagrams are also emerging that support larger data structures allowing more complex discrete representations to be implemented that was previously impossible.

This paper studies a class of renewable resource allocation problems for the processing of dynamically arriving tasks with deterministic deadlines. This class of problems has many applications, however, it conforms to neither the standard resource allocation model nor the standard optimal control model. A new problem formulation has to be developed and analyzed to provide a satisfactory solution. In this paper, a new formulation is presented. By augmenting the state variabls, the problem is converted into a Markovian decision problem. It can then be treated, at least in principle, by using the stochastic dynamic programming (SDP) method. However, since the system dynamics involves the evolution of sets (of tasks and resources), the implementation of the dynamic programming equation is by no means straightforward. For a problem with infinite planning horizon, the optimal strategy is shown to be stationary under mild conditions. An SDP algorithm based on a successive approximation technique is developed to obtain the optimal stationary strategy. The implementation of the algorithm employs a special coding scheme to handle set variables, and utilizes a dominance property for computational efficiency. Effects of key system parameters on optimal decisions are investigated and analyzed through numerical examples. As the computational complexity of the algorithm is of exponential increase, practical applications of the algorithm is limited to problems of moderate size. Two heuristic rules are therefore investigated and compared to the optimal policy.

We have developed a new derivative-free algorithm based on Radial Basis Functions (RBFs). Derivative-free optimization is an active field of research and several algorithms have been proposed recently. Problems of this nature in the industrial setting are quite frequent. The reason is that in a number of applications the optimization process contains simulation packages which are treated as black boxes. The development of our own algorithm was originally motivated by an application in biomedical imaging: the medical image registration problem. The particular characteristics of this problem have incited us to develop a new optimization algorithm based on trust-region methods. However it has been designed to be generic and to be applied to a wide range of problems. The main originality of our approach is the use of RBFs to build the models. In particular we have adapted the existing theory based on quadratic models to our own models and developed new procedures especially designed for models based on RBFs. We have tested our algorithm called BOOSTERS against state-of-the-art methods (UOBYQA, NEWUOA, DFO). On the medical image registration problem, BOOSTERS appears to be the method of choice. The tests on problems from the CUTEr collection show that BOOSTERS is comparable to, but not better than other methods on small problems (size 2-20). It is performing very well for medium size problems (20-80). Moreover, it is able to solve problems of dimension 200, which is considered very large in derivative-free optimization. We have also developed a new class of algorithms combining the robustness of derivative-free algorithms with the faster rate of convergence characterizing Newtonlike-methods. In fact, they define a new class of algorithms lying between derivative-free optimization and quasi-Newton methods. These algorithms are built on the skeleton of our derivative-free algorithm but they can incorporate the gradient when it is available. They can be interpreted as a way of doping derivative-free algorithms with derivatives. If the derivatives are available at each iteration, then our method can be seen as an alternative to quasi-Newton methods. At the opposite, if the derivatives are never evaluated, then the algorithm is totally similar to BOOSTERS. It is a very interesting alternative to existing methods for problems whose objective function is expensive to evaluate and when the derivatives are not available. In this situation, the gradient can be approximated by finite differences and its costs corresponds to n additional function evaluations assuming that Rn is the domain of definition of the objective function. We have compared our method with CFSQP and BTRA, two gradient-based algorithms, and the results show that our doped method performs best. We have also a theoretical analysis of the medical image registration problem based on maximization of mutual information. Most of the current research in this field is concentrated on registration based on nonlinear image transformation. However, little attention has been paid to the theoretical properties of the optimization problem. In our analysis, we focus on the continuity and the differentiability of the objective function. We show in particular that performing a registration without extension of the reference image may lead to discontinuities in the objective function. But we demonstrate that, under some mild assumptions, the function is differentiable almost everywhere. Our analysis is important from an optimization point of view and conditions the choice of a solver. The usual practice is to use generic optimization packages without worrying about the differentiability of the objective function. But the use of gradient-based methods when the objective function is not differentiable may result in poor performance or even in absence of convergence. One of our objectives with this analysis is also that practitioners become aware of these problems and to propose them new algorithms having a potential interest for their applications.

We address the multicast issue in a wide area WDM network in which a source broadcasts its message to some sites in the network. This problem can be formalized as the following optimal multicast tree problem. Given a directed graph G = (V, E) with a source s and a set S of nodes, assume that every node in S (⊂ V) is reachable from s. Associated with every link e ∈ E, there is a set Λ(e) of available wavelengths on it, |V| = n and |E| = m, find a multicast tree rooted at s including all the nodes in S with minimizing the cost of the tree, in terms of the sum of the cost of wavelength conversion at nodes and the cost of using wavelengths on links in the tree. That is, not only do we need to find a directed tree rooted at s, but also do we need to assign a specific wavelength Λ ∈ Λ(e) to every tree edge e and to set the switches at every node in the tree. Since this problem is NP-complete, it is unlikely that there is a polynomial algorithm to find an exact solution for it. Instead, we first present an approximation algorithm for the problem which runs in time O(k2n + km +kn log(kn) + (kn)1/e|S|), and delivers a solution within O(|S|e) of the optimal, for any fixed 0 > e ≤ 1. We then present a distributed version of the proposed algorithm. The communication and time complexities of the distributed algorithm are O(km) and O(kn) respectively, and the solution delivered is |S| times of the optimal, where k is the number of wavelengths in the network. Finally, we conduct some experiments to verify the applicability of the proposed algorithms. The experimental results matches our algorithmic claims.

Finding a delay constrained least-cost path in a directed graph is a fundamental issue for quality of service provisioning in communication networks. Recently Feng and Turkay proposed a hybrid algorithm which uses a heuristic approach to boost the average performance of a well known approximation algorithm. In this paper we provide extensive simulation results to show that the hybrid algorithm runs significantly faster than the approximation algorithm, often by one or two orders of magnitude.

Summary Random noise perturbs objective functions in a variety of practical optimization problems, and genetic algorithms (GAs) have been widely proposed as an effective optimization tool for dealing with noisy objective functions. In this paper, we investigate GAs applied to objective functions that are perturbed by multiple sources of additive and multiplicative noise that each take on finitely many values. We reveal the convergence properties of GAs by constructing and analyzing a Markov chain that explicitly models the evolution of the algorithms in noisy environments. Our analysis shows that this Markov chain is indecomposable; it has only one positive recurrent communication class. Using this property, we establish a condition that is both necessary and sufficient for GAs to eventually (i.e., as the number of iterations goes to infinity) find a globally optimal solution with probability 1. Similarly, we identify a condition that is both necessary and sufficient for the algorithms to eventually with probability 1 fail to find any globally optimal solution. We also discuss how the transition probabilities of the chain can be used to derive an upper bound for the number of iterations sufficient to ensure with certain probability that a GA selects a globally optimal solution upon termination.

In this thesis we analyze and develop methods based on model order reduction (MOR) for the solution of optimization problems constrained by time-dependent partial differential equations (PDEs). The methods combine a direct solution approach with model reduction via proper orthogonal decomposition (POD) and the discrete empirical interpolation method (DEIM). The reduced-order models (ROMs) are used to approximate the high-dimensional dynamic systems originating from a spatial discretization of a PDE. However, when used in an optimization algorithm, conventional POD/DEIM ROMs often lack the ability to give adequate approximations of the gradient. We propose methods for a suitable enhancement of the ROMs for the optimization purpose which are based on the inclusion of derivative information in the POD and DEIM subspaces. We distinguish two types of error between quantities evaluated with the high-dimensional model and its ROM approximation in dependency on the optimization variable q: The reconstruction error which is evaluated with the same q0 which is used constructing the ROM and the prediction error which assesses approximations at q with a ROM constructed at q0 different to q. The novel reconstruction results we present include estimates for solutions of the adjoint equation and the sensitivity equations as well as for the gradient of the objective function. Based on the estimates we explain how the POD and DEIM bases should be extended with either adjoint or sensitivity information. The enhanced ROMs allow control of the reconstruction error for the objective and its gradient up to machine precision. Moreover, we propose a POD prediction estimate for the objective of the optimization problem in a neighborhood of q where the ROM is constructed. In case of sensitivity-extended POD and DEIM bases we give an analogous result for solutions of the states. The derivative-extended ROMs are then used to develop adaptive algorithms for the solution of optimal control and parameter estimation problems which results in great runtime improvements for the optimization while ensuring high approximation quality of the solution of the original problem. For the parameter estimation case a novel a posteriori error estimate is proposed which assesses the quality of suboptimal solutions obtained with the ROM. A further fundamental contribution is a discussion of discretize-then-optimize (DTO) vs. optimize-then-discretize (OTD) approaches in the context of MOR for optimization. We analyze advantages and disadvantages of both approaches and discuss to which extent our methods exhibit properties of either strategy. We also give examples of representative optimization problems in which standard POD/DEIM ROMs show an inacceptable behavior and can be successfully solved by derivative-extended ROMs. We have further implemented the developed methods emphasizing an efficient realization which is important for the investigation of the MOR potential. We showcase the practical performance of the proposed algorithms and the superiority of derivative-extended over conventional ROMs on two academic and one industry-relevant application which exhibit a variety of challenges for the model reduction approach in optimization.

In the existing quantum-behaved optimization algorithms, almost all of the individuals are encoded by qubits described on plane unit circle. As qubits contain only a variable parameter, quantum properties have not been fully embodied, which limits the optimization ability rise further. In order to solve this problem, this paper proposes a quantum ant colony optimization algorithm based on Bloch sphere search. In the proposed algorithm, the positions of ants are encoded by qubits described on Bloch sphere. First, the destination to move is determined according to the select probability constructed by the pheromone and heuristic information, then, the rotation axis is established with Pauli matrixes, and the evolution search is realized with the rotation of qubits on Bloch sphere. In order to avoid premature convergence, the mutation is performed with Hadamard gates. Finally, the pheromone and the heuristic information are updated in the new positions of ants. As the optimization process is performed in n-dimensional hypercube space (−1, 1) n , which has nothing to do with the specific issues, hence, the proposed method has good adaptability for a variety of optimization problems. The simulation results show that the proposed algorithm is superior to other quantum-behaved optimization algorithms in both search ability and optimization efficiency.

Storage Logistics of Fruits and Vegetables in Distribution Centers

ABSTRACTThe cost of nitrogen storage to current growth was examined in relation to N availability in the biennial Cirsium vulgare. Plants were grown outdoors, in sand culture, with continuous diel drip irrigation of fertilization medium containing one of five different N concentrations. Plants grown at the highest N concentration stored twice as much N in their tap roots as did plants grown at the lowest N concentration. In high‐N‐grown plants, the storage of N reserves occurred during the period of maximum growth, at the same time as tap‐root production. At the time of maximum biomass, stored N was also at a maximum. During the period following maximum biomass, no additional storage of N occurred. This pattern was observed despite frequent late‐season leaf senescence which resulted in a large pool of potentially mobile N which could have been stored at no cost to growth. In low‐N‐grown plants, the production of tap‐root storage tissue and the filling of that tissue with stored N were staggered. Tap‐root production and growth occurred during the period of maximum growth, as in the high‐N‐grown plants. However, filling of the storage tissue with N occurred late in the growing season, when the pool of mobile N from senescent leaves was large. The utilization of this late‐season N source occurred with little or no cost to growth, and this N is labelled, according to previous definitions, as ‘accumulated’. The costs of storing N in plants of the different N treatments were calculated using two models based on different growth constraints. In one model, the cost of N storage was represented as lost growth due to allocation of N to storage, rather than to the photosynthetic shoot (i.e. growth was assumed to be limited by carbon acquisition). In the second model, the storage cost was calculated as lost growth due to allocation of N to storage, rather than to the nitrogen‐acquiring fine‐root system (i.e. growth was assumed to be limited by nitrogen acquisition). In both models, the total cost of N storage was predicted to decrease as N availability decreased due to smaller storage pool sizes in plants of the low‐N treatments. The cost of filling the tap root with stored N as a percentage of the total storage cost was also reduced as N availability decreased due to the occurrence of late‐season accumulation. By relying, at least in part, on late‐season accumulation, plants grown at the lowest three levels of N availability reduced total storage costs by 15 to 22%. The results demonstrate that plants are capable of adjusting their storage patterns in response to low nitrogen availability such that the costs of storage are reduced.

This thesis focuses on the development and analysis of general methods for the design of optimal discrete PID control strategies for multivariable glass furnace processes, where standard genetic algorithms (SGAs) are applied to optimise specially formulated objective functions. Furthermore, a strong emphasis is given on the realistic model parameters identi cation method, which is illustrated to be applicable to a wide range of higher order model parameters identi cation problems. A complete, realistic and continuous excess oxygen model with nonlinearity effect was developed and the model parameters were identified. The developed excess oxygen model consisted of three sub-models to characterise the real plant response. The developed excess oxygen model was evaluated and compared with real plant dynamic response data, which illustrated the high degree of accuracy of the developed model. A new technique named predetermined time constant approximation was proposed to make an assumption on the initial value of a predetermined time constant, whose motive is to facilitate the SGAs to explore and exploit an optimal value for higher order of continuous model's parameters identi cation. Also, the proposed predetermined time constant approximation technique demonstrated that the population diversity is well sustained while exploring the feasible search region and exploiting to an optimal value. In general, the proposed method improves the SGAs convergence rate towards the global optimum and illustrated the effectiveness. An automatic tuning of decentralised discrete PID controllers for multivariable processes, based on SGAs, was proposed. The main improvement of the proposed technique is the ability to enhance the control robustness and to optimise discrete PID parameters by compensating the loop interaction of a multivariable process. This is attained by adding the individually optimised objective function of glass temperature and excess oxygen processes as one objective function, to include the total effect of the loop interaction by applying step inputs on both set points, temperature and excess oxygen, at two different time periods in one simulation. The effectiveness of the proposed tuning technique was supported by a number of simulation results using two other SGAs conventional tuning techniques with 1st and 2nd order control oriented models. It was illustrated that, in all cases, the resulting discrete PID control parameters completely satisfied all performance specifications. A new technique to minimise the fuel consumption for glass furnace processes while sustaining the glass temperature is proposed. This proposed technique is achieved by reducing the excess oxygen within the optimum thermal efficiency region within 1.7% to 3.2%, which is approximately equal to about 10% to 20% of excess air. Therefore, by reducing the excess oxygen set point within the optimum region, 2.45% to 2%, the fuel consumption is minimised from 0:002942kg/sec to 0:002868kg/sec while the thermal efficiency of the glass temperature is sustained at the desired set point (1550K). In addition, a reduction in excess oxygen within methane combustion guidelines will assure that undesirable emissions are in control throughout the combustion process. The efficiencies of the proposed technique were supported by a number of simulation results applying the three SGAs controller tuning techniques. It was illustrated that, in all cases, the fraction of excess oxygen reduction results in a great minimisation of fuel consumption over long plant operating periods.

In this paper we describe a new technique for placing fixed capacitors in radial distribution systems based on genetic algorithms (GA). Actually, current optimization models of capacitor placement only consider losses reduction and voltage profile simultaneously, but the compensation cost and the load changes are not taking into account as part of the objective function. Also, the result may be not the best choice because this is a very large optimization problem and there are too many combinations. We present here a general approach for the optimal solution of this problem considering all the parameters of the distribution system involved: capacitor cost, voltage, angle and load changes per hour. An exhaustive search through all possible solutions is needed. Therefore, GA is an ideal candidate to solve this situation. Working with this algorithm, a Microsoft Excel Program has been developed in order to answer the capacitor placement problem for any kind of radial distribution system. The test results are presented along with the discussion of the algorithm.

Reinforcement learning (RL) focuses on an essential aspect of intelligent behavior – how an agent can learn to make good decisions given experience and rewards in a stochastic<br>world. Yet popular RL algorithms that have enabled exciting successes in domains with good simulators (Go, Atari, etc) still often fail to learn in other domains because they rely on<br>simple heuristics for exploration. This provides additional empirical justification for essential questions around RL, specifically around algorithms that learn in a provably efficient manner through strategic exploration in any considered domain. This thesis provides new algorithms<br>and theory that enable good performance with respect to existing theoretical frameworks for evaluating RL algorithms (specifically, probably approximately correct) and introduces new stronger evaluation criteria, that may be particularly of interest as RL is applied to more real world problems.<br>For the first line of work on probably approximately correct (PAC) RL algorithms, we introduce a series of algorithms for episodic tabular domains with substantially better PAC<br>sample complexity bounds that culminate in a new algorithm with close to minimax optimal PAC and regret bounds. Look up tables are required by most sample efficient and computationally tractable algorithms, but cannot represent many practical domains. We therefore also present a new RL algorithm that can learn a good policy in environments with high dimensional observations and hidden deterministic states; unlike predecessors, this algorithm provably<br>explores not only in a statistically but also computationally efficient manner assuming access to function classes with efficient optimization oracles. To make progress it is critical to have the right measures of success. While empirical<br>demonstrations are quite clear, we find that for theoretical properties, two of the most commonly used learning frameworks, PAC guarantees and regret guarantees, each allow undesirable algorithm behavior (e.g. ignoring new observations that could improve the policy). We present<br>a new stronger learning framework called Uniform-PAC that unifies the existing frameworks and prevents undesirable algorithm properties. One caveat of all existing learning frameworks is that for any particular episode, we do not<br>know how well the algorithm will perform. To address this, we introduce the IPOC framework that requires algorithms to provide a certificate before each episode bounding how suboptimal the current policy can be. Such certifications may be of substantial interest in high stakes scenarios when an organization may wish to track or even pause an online RL system should the potential expected performance bound drop below a required expected outcome.

This work describes the optimization process of a morphing wing for an unmanned air vehicle (UAV) by using multi-objective gradient based optimization algorithm in which maximization of lift to drag ratio and minimization of bending moment at wing root chord plane are defined as the objective functions. Six different optimization problems are solved by changing weighted sum coefficients of the objective functions in order to observe the morphing wing planform and the airfoil shape change. According to the results, 15.2% lift to drag ratio increase is obtained while the increase in the bending moment is 2%.

This paper proposes a color feature description for image and video retrieving applications such as personal video recorder, which manages a large amount of data. This descriptor specifies the spatial distribution of colors with a few nonlinear quantized DCT coefficients of grid based average colors. The tradeoff between the number of the coefficients enclosed in the descriptor and retrieval cost is studied. The experimental results show that the descriptor enclosing six for luminance and three for each chrominance coefficient achieves the best trade-off between the storage cost and retrieval efficiency. It requires 6 bits for DC and 5 bits for AC coefficients, and therefore total storage cost is just 63 bits per image. This description, named color layout descriptor, has been accepted as a part of the MPEG-7 final committee draft.