Three dimensional Multiprocessor System-on-Chip (3D-MPSoC) adoption. It is characterized by the integration of a large amount of hardware components on a single multilayer chip. However, heating is one of the major pi...
详细信息
ISBN:
(纸本)9781479911301
Three dimensional Multiprocessor System-on-Chip (3D-MPSoC) adoption. It is characterized by the integration of a large amount of hardware components on a single multilayer chip. However, heating is one of the major pitfalls of the 3D-MPSoCs. Three dimensional Network-on-Chip (3D-NoC) is used as the communication structure of 3D-MPSoCs. Its main role in system operation and performance makes the optimal 3D-NoC design a critical task. Final 3D-NoC configuration must fulfill all the application requirements and heating constraints of the system. Topology and mapping are some of the most critical parameters in 3D-NoC design, strongly influencing the 3D-MPSoC performance and cost. 3D-NoC topology and mapping has been solved for single application systems on homogeneous 3D-NoCs using single and multi-objective optimization algorithms. In this paper we use a multi-objective immune algorithm (MIA), to solve the multi-application 3D-NoC topology and mapping problems. Latency and power consumption are adopted as the target multi-objective functions constrained by the heating function. Our strategy has been applied on 8 3D-MPSoC benchmarks. Their final 3D-NoC configurations have up to 73% power and 42% latency enhancement when compared to previous reported results.
We investigate three related and important problems connected to machine learning: approximating a submodular function everywhere, learning a submodular function (in a PAC-like setting), and constrained minimization o...
详细信息
ISBN:
(纸本)9781632660244
We investigate three related and important problems connected to machine learning: approximating a submodular function everywhere, learning a submodular function (in a PAC-like setting), and constrained minimization of submodular functions. We show that the complexity of all three problems depends on the "curvature" of the submodular function, and provide lower and upper bounds that refine and improve previous results. Our proof techniques are fairly generic. We either use a black-box transformation of the function (for approximation and learning), or a transformation of algorithms to use an appropriate surrogate function (for minimization). Curiously, curvature has been known to influence approximations for submodular maximization, but its effect on minimization, approximation and learning has hitherto been open. We complete this picture, and also support our theoretical claims by empirical results.
In this paper we propose a model predictive control scheme for discrete-time linear time-invariant systems based on inexact numerical optimization algorithms. We assume that the solution of the associated quadratic pr...
详细信息
ISBN:
(纸本)9781467357159
In this paper we propose a model predictive control scheme for discrete-time linear time-invariant systems based on inexact numerical optimization algorithms. We assume that the solution of the associated quadratic program produced by some numerical algorithm is possibly neither optimal nor feasible, but the algorithm is able to provide estimates on primal suboptimality and primal feasibility violation. By tightening the complicating constraints we can ensure the primal feasibility of the approximate solutions generated by the algorithm. Finally, we derive a control strategy that has the following properties: the constraints on the states and inputs are satisfied, asymptotic stability of the closed-loop system is guaranteed, and the number of iterations needed for a desired level of suboptimality can be determined.
Most distributed optimization methods used for distributed model predictive control (DMPC) are gradient based. Gradient based optimization algorithms are known to have iterations of low complexity. However, the number...
详细信息
ISBN:
(纸本)9781479901777
Most distributed optimization methods used for distributed model predictive control (DMPC) are gradient based. Gradient based optimization algorithms are known to have iterations of low complexity. However, the number of iterations needed to achieve satisfactory accuracy might be significant. This is not a desirable characteristic for distributed optimization in distributed model predictive control. Rather, the number of iterations should be kept low to reduce communication requirements, while the complexity within an iteration can be significant. By incorporating Hessian information in a distributed accelerated gradient method in a well-defined manner, we are able to significantly reduce the number of iterations needed to achieve satisfactory accuracy in the solutions, compared to distributed methods that are strictly gradient-based. Further, we provide convergence rate results and iteration complexity bounds for the developed algorithm.
Recently Raghavendra and Tan (SODA 2012) gave a 0.85 approximation algorithm for the Max Bisection problem. We improve their algorithm to a 0.8776-approximation. As Max Bisection is hard to approximate within α_(GW)...
详细信息
ISBN:
(纸本)9781611972511
Recently Raghavendra and Tan (SODA 2012) gave a 0.85 approximation algorithm for the Max Bisection problem. We improve their algorithm to a 0.8776-approximation. As Max Bisection is hard to approximate within α_(GW) + ε ≈ 0.8786 under the Unique Games Conjecture (UGC), our algorithm is nearly optimal. We conjecture that Max Bisection is approximable within α_(GW)-ε, i.e., the bisection constraint (essentially) does not make Max Cut harder. We also obtain an optimal algorithm (assuming the UGC) for the analogous variant of MAX 2-Sat. Our approximation ratio for this problem exactly matches the optimal approximation ratio for MAX 2-Sat, i.e., α_(LLZ)+ε≈ 0.9401, showing that the bisection constraint does not make MAX 2-Sat harder. This improves on a 0:93-approximation for this problem due to Raghavendra and Tan.
We present and study a distributed optimization algorithm by employing a stochastic dual coordinate ascent method. Stochastic dual coordinate ascent methods enjoy strong theoretical guarantees and often have better pe...
详细信息
ISBN:
(纸本)9781632660244
We present and study a distributed optimization algorithm by employing a stochastic dual coordinate ascent method. Stochastic dual coordinate ascent methods enjoy strong theoretical guarantees and often have better performances than stochastic gradient descent methods in optimizing regularized loss minimization problems. It still lacks of efforts in studying them in a distributed framework. We make a progress along the line by presenting a distributed stochastic dual coordinate ascent algorithm in a star network, with an analysis of the tradeoff between computation and communication. We verify our analysis by experiments on real data sets. Moreover, we compare the proposed algorithm with distributed stochastic gradient descent methods and distributed alternating direction methods of multipliers for optimizing SVMs in the same distributed framework, and observe competitive performances.
A scalable multiple description scalar quantizer (SMDSQ) is a quantization based framework used for scalable multiple description coding (SMDC). In this paper, we introduce a novel generalization of the Lloyd-Max algo...
详细信息
ISBN:
(纸本)9784885522673
A scalable multiple description scalar quantizer (SMDSQ) is a quantization based framework used for scalable multiple description coding (SMDC). In this paper, we introduce a novel generalization of the Lloyd-Max algorithm to realize locally optimal SMDSQs. Both level-constrained and entropy-constrained cases are considered. For both cases, locally optimal solutions are realized by iterative execution of the centroid and the modified nearest-neighbor conditions. Experimental results confirm that, for a zero-mean unit-variance Gaussian source, the optimization algorithm enables a significant reduction in distortion for the level-constrained case. Moreover, relatively lesser but still significant distortion-rate (D-R) gains are viable for the entropy-constrained case. It is shown that, for a packetized transmission of Gaussian as well as wavelet-decomposed images, the obtained optimization gains translate into an average improvement in the decoder's signal-to-noise-ratio (SNR) for a wide range of packet loss rates.
This paper aims to optimize the design of a novel antenna for aerospace applications to be integrated on an experimental rocket that has been designed in an advanced research student program. In order to optimize EM p...
详细信息
ISBN:
(纸本)9781467357067
This paper aims to optimize the design of a novel antenna for aerospace applications to be integrated on an experimental rocket that has been designed in an advanced research student program. In order to optimize EM performance of such a system a novel optimization algorithm called SNO, Social Network optimization, has been developed and tested to find the best configuration that assure the effectiveness of the overall rocket design. The explosion of Internet and Social Network diffusion in the last years is having a direct impact on different research fields as computer science, economy, sociology and biological science. SNO concept is an optimization algorithm inspired by the emulation of decision making process in social network environments. The experimental results of the rocket antenna design will be presented to prove the effectiveness of this new algorithm.
There has been a great interest in integration of distributed generation (DG) units at distribution level in the recent years. DGs can provide cost-effective, environmental friendly, higher power quality and more reli...
详细信息
There has been a great interest in integration of distributed generation (DG) units at distribution level in the recent years. DGs can provide cost-effective, environmental friendly, higher power quality and more reliable energy solutions than conventional generation. For maximum power loss reduction, proper sizing and position of distributed generators are ardently necessary. This paper presents a simple method for optimizing cost and optimal placement of generators. A simple vector based load flow technique is implemented on 38 bus distribution systems. This paper presents a new methodology using a new population based meta heuristic approach namely Shuffled frog leaping algorithm for the placement of Distributed Generators (DG) in the radial distribution systems to reduce the real power losses and cost of the DG. The paper also focuses on optimization of weighting factor, which balances the cost and the loss factors and helps to build up desired objectives with maximum potential benefit. The proposed method has outperformed the other methods in terms of the quality of solution and computational efficiency. (C) 2011 Published by Elsevier Ltd. Selection and/or peer-review under responsibility of the organizing committee of 2nd International Conference on Advances in Energy Engineering (ICAEE).
Controlling the position of elements in sparse planar array would be the hardest work in the synthesis procedure since the array has to satisfy multiple design constraints e.g. number of elements, elements spacing and...
详细信息
ISBN:
(纸本)9781467357067
Controlling the position of elements in sparse planar array would be the hardest work in the synthesis procedure since the array has to satisfy multiple design constraints e.g. number of elements, elements spacing and arrays aperture. In this paper, a simple method is implemented to effectively control 2-D sparse array. The approach implements two sets of data to separately manage the x- and y-coordinate of each element in the array. The array is then synthesized as an optimization problem using the recently improved version of Bayesian optimization Algorithm. As a proof of concept, the results of a 108 element sparse planar array are here presented and discussed.
暂无评论