PyTorchKR Papers

최신 논문

총 605편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:1906.10760

Distributed Optimization for Smart Cyber-Physical Networks

The presence of embedded electronics and communication capabilities as well as sensing and control in smart devices has given rise to the novel concept of cyber-physical networks, in which agents aim at cooperatively solving complex tasks by local computation and communication. Numerous estimation, learning, decision and control tasks in smart networks involve the solution of large-scale, structured optimization problems in which network agents have only a partial knowledge of the whole problem. Distributed optimization aims at designing local computation and communication rules for the network processors allowing them to cooperatively solve the global optimization problem without relying on any central unit. The purpose of this survey is to provide an introduction to distributed optimization methodologies. Principal approaches, namely (primal) consensus-based, duality-based and constraint exchange methods, are formalized. An analysis of the basic schemes is supplied, and state-of-the-art extensions are reviewed.

arXiv:2010.10698

Batch Sequential Adaptive Designs for Global Optimization

Compared with the fixed-run designs, the sequential adaptive designs (SAD) are thought to be more efficient and effective. Efficient global optimization (EGO) is one of the most popular SAD methods for expensive black-box optimization problems. A well-recognized weakness of the original EGO in complex computer experiments is that it is serial, and hence the modern parallel computing techniques cannot be utilized to speed up the running of simulator experiments. For those multiple points EGO methods, the heavy computation and points clustering are the obstacles. In this work, a novel batch SAD method, named "accelerated EGO", is forwarded by using a refined sampling/importance resampling (SIR) method to search the points with large expected improvement (EI) values. The computation burden of the new method is much lighter, and the points clustering is also avoided. The efficiency of the proposed SAD is validated by nine classic test functions with dimension from 2 to 12. The empirical results show that the proposed algorithm indeed can parallelize original EGO, and gain much improvement compared against the other parallel EGO algorithm especially under high-dimensional case. Additionally, we also apply the new method to the hyper-parameter tuning of Support Vector Machine (SVM). Accelerated EGO obtains comparable cross validation accuracy with other methods and the CPU time can be reduced a lot due to the parallel computation and sampling method.

arXiv:2010.05343

Non-Stationary Stochastic Global Optimization Algorithms

Gomez proposes a formal and systematic approach for characterizing stochastic global optimization algorithms. Using it, Gomez formalizes algorithms with a fixed next-population stochastic method, i.e., algorithms defined as stationary Markov processes. These are the cases of standard versions of hill-climbing, parallel hill-climbing, generational genetic, steady-state genetic, and differential evolution algorithms. This paper continues such a systematic formal approach. First, we generalize the sufficient conditions convergence lemma from stationary to non-stationary Markov processes. Second, we develop Markov kernels for some selection schemes. Finally, we formalize both simulated-annealing and evolutionary-strategies using the systematic formal approach.

arXiv:2010.03171공식 코드 · Papers with Code

Additive Tree-Structured Conditional Parameter Spaces in Bayesian Optimization: A Novel Covariance Function and a Fast Implementation

Bayesian optimization (BO) is a sample-efficient global optimization algorithm for black-box functions which are expensive to evaluate. Existing literature on model based optimization in conditional parameter spaces are usually built on trees. In this work, we generalize the additive assumption to tree-structured functions and propose an additive tree-structured covariance function, showing improved sample-efficiency, wider applicability and greater flexibility. Furthermore, by incorporating the structure information of parameter spaces and the additive assumption in the BO loop, we develop a parallel algorithm to optimize the acquisition function and this optimization can be performed in a low dimensional space. We demonstrate our method on an optimization benchmark function, on a neural network compression problem, on pruning pre-trained VGG16 and ResNet50 models as well as on searching activation functions of ResNet20. Experimental results show our approach significantly outperforms the current state of the art for conditional parameter optimization including SMAC, TPE and Jenatton et al. (2017).

arXiv:2010.03258

Global Optimization of Objective Functions Represented by ReLU Networks

Neural networks can learn complex, non-convex functions, and it is challenging to guarantee their correct behavior in safety-critical contexts. Many approaches exist to find failures in networks (e.g., adversarial examples), but these cannot guarantee the absence of failures. Verification algorithms address this need and provide formal guarantees about a neural network by answering "yes or no" questions. For example, they can answer whether a violation exists within certain bounds. However, individual "yes or no" questions cannot answer qualitative questions such as "what is the largest error within these bounds"; the answers to these lie in the domain of optimization. Therefore, we propose strategies to extend existing verifiers to perform optimization and find: (i) the most extreme failure in a given input region and (ii) the minimum input perturbation required to cause a failure. A naive approach using a bisection search with an off-the-shelf verifier results in many expensive and overlapping calls to the verifier. Instead, we propose an approach that tightly integrates the optimization process into the verification procedure, achieving better runtime performance than the naive approach. We evaluate our approach implemented as an extension of Marabou, a state-of-the-art neural network verifier, and compare its performance with the bisection approach and MIPVerify, an optimization-based verifier. We observe complementary performance between our extension of Marabou and MIPVerify.

arXiv:2010.01524

Multi-Level Evolution Strategies for High-Resolution Black-Box Control

This paper introduces a multi-level (m-lev) mechanism into Evolution Strategies (ESs) in order to address a class of global optimization problems that could benefit from fine discretization of their decision variables. Such problems arise in engineering and scientific applications, which possess a multi-resolution control nature, and thus may be formulated either by means of low-resolution variants (providing coarser approximations with presumably lower accuracy for the general problem) or by high-resolution controls. A particular scientific application concerns practical Quantum Control (QC) problems, whose targeted optimal controls may be discretized to increasingly higher resolution, which in turn carries the potential to obtain better control yields. However, state-of-the-art derivative-free optimization heuristics for high-resolution formulations nominally call for an impractically large number of objective function calls. Therefore, an effective algorithmic treatment for such problems is needed. We introduce a framework with an automated scheme to facilitate guided-search over increasingly finer levels of control resolution for the optimization problem, whose on-the-fly learned parameters require careful adaptation. We instantiate the proposed m-lev self-adaptive ES framework by two specific strategies, namely the classical elitist single-child (1+1)-ES and the non-elitist multi-child derandomized $(\mu_W,\lambda)$-sep-CMA-ES. We first show that the approach is suitable by simulation-based optimization of QC systems which were heretofore viewed as too complex to address. We also present a laboratory proof-of-concept for the proposed approach on a basic experimental QC system objective.

arXiv:2010.01319

Deep learning algorithms for solving high dimensional nonlinear backward stochastic differential equations

In this work, we propose a new deep learning-based scheme for solving high dimensional nonlinear backward stochastic differential equations (BSDEs). The idea is to reformulate the problem as a global optimization, where the local loss functions are included. Essentially, we approximate the unknown solution of a BSDE using a deep neural network and its gradient with automatic differentiation. The approximations are performed by globally minimizing the quadratic local loss function defined at each time step, which always includes the terminal condition. This kind of loss functions are obtained by iterating the Euler discretization of the time integrals with the terminal condition. Our formulation can prompt the stochastic gradient descent algorithm not only to take the accuracy at each time layer into account, but also converge to a good local minima. In order to demonstrate performances of our algorithm, several high-dimensional nonlinear BSDEs including pricing problems in finance are provided.

arXiv:2004.12061공식 코드 · Papers with Code

Nonlinear Dynamic Systems Parameterization Using Interval-Based Global Optimization: Computing Lipschitz Constants and Beyond

Numerous state-feedback and observer designs for nonlinear dynamic systems (NDS) have been developed in the past three decades. These designs assume that NDS nonlinearities satisfy one of the following function set classifications: bounded Jacobian, Lipschitz continuity, one-sided Lipschitz, quadratic inner-boundedness, and quadratic boundedness. These function sets are characterized by constant scalars or matrices bounding the NDS' nonlinearities. These constants (i) depend on the NDS' operating region, topology, and parameters, and (ii) are utilized to synthesize observer/controller gains. Unfortunately, there is a near-complete absence of algorithms to compute such bounding constants. In this paper, we develop analytical then computational methods to compute such constants. First, for every function set classification, we derive analytical expressions for these bounding constants through global maximization formulations. Second, we utilize a derivative-free, interval-based global maximization algorithm based on branch-and-bound framework to numerically obtain the bounding constants. Third, we showcase the effectiveness of our approaches to compute the corresponding parameters on some NDS such as highway traffic networks and synchronous generator models.

arXiv:2009.13772

Trust-Region Method with Deep Reinforcement Learning in Analog Design Space Exploration

This paper introduces new perspectives on analog design space search. To minimize the time-to-market, this endeavor better cast as constraint satisfaction problem than global optimization defined in prior arts. We incorporate model-based agents, contrasted with model-free learning, to implement a trust-region strategy. As such, simple feed-forward networks can be trained with supervised learning, where the convergence is relatively trivial. Experiment results demonstrate orders of magnitude improvement on search iterations. Additionally, the unprecedented consideration of PVT conditions are accommodated. On circuits with TSMC 5/6nm process, our method achieve performance surpassing human designers. Furthermore, this framework is in production in industrial settings.

arXiv:2009.03586공식 코드 · Papers with Code

Hyperparameter Optimization via Sequential Uniform Designs

Hyperparameter optimization (HPO) plays a central role in the automated machine learning (AutoML). It is a challenging task as the response surfaces of hyperparameters are generally unknown, hence essentially a global optimization problem. This paper reformulates HPO as a computer experiment and proposes a novel sequential uniform design (SeqUD) strategy with three-fold advantages: a) the hyperparameter space is adaptively explored with evenly spread design points, without the need of expensive meta-modeling and acquisition optimization; b) the batch-by-batch design points are sequentially generated with parallel processing support; c) a new augmented uniform design algorithm is developed for the efficient real-time generation of follow-up design points. Extensive experiments are conducted on both global optimization tasks and HPO applications. The numerical results show that the proposed SeqUD strategy outperforms benchmark HPO methods, and it can be therefore a promising and competitive alternative to existing AutoML tools.

arXiv:2009.02560

Particle Swarm Optimized Federated Learning For Industrial IoT and Smart City Services

Most of the research on Federated Learning (FL) has focused on analyzing global optimization, privacy, and communication, with limited attention focusing on analyzing the critical matter of performing efficient local training and inference at the edge devices. One of the main challenges for successful and efficient training and inference on edge devices is the careful selection of parameters to build local Machine Learning (ML) models. To this aim, we propose a Particle Swarm Optimization (PSO)-based technique to optimize the hyperparameter settings for the local ML models in an FL environment. We evaluate the performance of our proposed technique using two case studies. First, we consider smart city services and use an experimental transportation dataset for traffic prediction as a proxy for this setting. Second, we consider Industrial IoT (IIoT) services and use the real-time telemetry dataset to predict the probability that a machine will fail shortly due to component failures. Our experiments indicate that PSO provides an efficient approach for tuning the hyperparameters of deep Long short-term memory (LSTM) models when compared to the grid search method. Our experiments illustrate that the number of clients-server communication rounds to explore the landscape of configurations to find the near-optimal parameters are greatly reduced (roughly by two orders of magnitude needing only 2%--4% of the rounds compared to state of the art non-PSO-based approaches). We also demonstrate that utilizing the proposed PSO-based technique to find the near-optimal configurations for FL and centralized learning models does not adversely affect the accuracy of the models.

arXiv:2009.02183

On the implementation of a global optimization method for mixed-variable problems

We describe the optimization algorithm implemented in the open-source derivative-free solver RBFOpt. The algorithm is based on the radial basis function method of Gutmann and the metric stochastic response surface method of Regis and Shoemaker. We propose several modifications aimed at generalizing and improving these two algorithms: (i) the use of an extended space to represent categorical variables in unary encoding; (ii) a refinement phase to locally improve a candidate solution; (iii) interpolation models without the unisolvence condition, to both help deal with categorical variables, and initiate the optimization before a uniquely determined model is possible; (iv) a master-worker framework to allow asynchronous objective function evaluations in parallel. Numerical experiments show the effectiveness of these ideas.

arXiv:2009.01272공식 코드 · Papers with Code

Understanding the wiring evolution in differentiable neural architecture search

Controversy exists on whether differentiable neural architecture search methods discover wiring topology effectively. To understand how wiring topology evolves, we study the underlying mechanism of several existing differentiable NAS frameworks. Our investigation is motivated by three observed searching patterns of differentiable NAS: 1) they search by growing instead of pruning; 2) wider networks are more preferred than deeper ones; 3) no edges are selected in bi-level optimization. To anatomize these phenomena, we propose a unified view on searching algorithms of existing frameworks, transferring the global optimization to local cost minimization. Based on this reformulation, we conduct empirical and theoretical analyses, revealing implicit inductive biases in the cost's assignment mechanism and evolution dynamics that cause the observed phenomena. These biases indicate strong discrimination towards certain topologies. To this end, we pose questions that future differentiable methods for neural wiring discovery need to confront, hoping to evoke a discussion and rethinking on how much bias has been enforced implicitly in existing NAS methods.

arXiv:2009.00935

Real-time 3D Facial Tracking via Cascaded Compositional Learning

We propose to learn a cascade of globally-optimized modular boosted ferns (GoMBF) to solve multi-modal facial motion regression for real-time 3D facial tracking from a monocular RGB camera. GoMBF is a deep composition of multiple regression models with each is a boosted ferns initially trained to predict partial motion parameters of the same modality, and then concatenated together via a global optimization step to form a singular strong boosted ferns that can effectively handle the whole regression target. It can explicitly cope with the modality variety in output variables, while manifesting increased fitting power and a faster learning speed comparing against the conventional boosted ferns. By further cascading a sequence of GoMBFs (GoMBF-Cascade) to regress facial motion parameters, we achieve competitive tracking performance on a variety of in-the-wild videos comparing to the state-of-the-art methods, which require much more training data or have higher computational complexity. It provides a robust and highly elegant solution to real-time 3D facial tracking using a small set of training data and hence makes it more practical in real-world applications.

arXiv:2008.07049

Video Region Annotation with Sparse Bounding Boxes

Video analysis has been moving towards more detailed interpretation (e.g. segmentation) with encouraging progresses. These tasks, however, increasingly rely on densely annotated training data both in space and time. Since such annotation is labour-intensive, few densely annotated video data with detailed region boundaries exist. This work aims to resolve this dilemma by learning to automatically generate region boundaries for all frames of a video from sparsely annotated bounding boxes of target regions. We achieve this with a Volumetric Graph Convolutional Network (VGCN), which learns to iteratively find keypoints on the region boundaries using the spatio-temporal volume of surrounding appearance and motion. The global optimization of VGCN makes it significantly stronger and generalize better than existing solutions. Experimental results using two latest datasets (one real and one synthetic), including ablation studies, demonstrate the effectiveness and superiority of our method.

arXiv:2008.07491

Metaheuristic optimization of power and energy systems: underlying principles and main issues of the 'rush to heuristics'

In the power and energy systems area, a progressive increase of literature contributions containing applications of metaheuristic algorithms is occurring. In many cases, these applications are merely aimed at proposing the testing of an existing metaheuristic algorithm on a specific problem, claiming that the proposed method is better than other methods based on weak comparisons. This 'rush to heuristics' does not happen in the evolutionary computation domain, where the rules for setting up rigorous comparisons are stricter, but are typical of the domains of application of the metaheuristics. This paper considers the applications to power and energy systems, and aims at providing a comprehensive view of the main issues concerning the use of metaheuristics for global optimization problems. A set of underlying principles that characterize the metaheuristic algorithms is presented. The customization of metaheuristic algorithms to fit the constraints of specific problems is discussed. Some weaknesses and pitfalls found in literature contributions are identified, and specific guidelines are provided on how to prepare sound contributions on the application of metaheuristic algorithms to specific problems.

arXiv:2008.02436

GL-GAN: Adaptive Global and Local Bilevel Optimization model of Image Generation

Although Generative Adversarial Networks have shown remarkable performance in image generation, there are some challenges in image realism and convergence speed. The results of some models display the imbalances of quality within a generated image, in which some defective parts appear compared with other regions. Different from general single global optimization methods, we introduce an adaptive global and local bilevel optimization model(GL-GAN). The model achieves the generation of high-resolution images in a complementary and promoting way, where global optimization is to optimize the whole images and local is only to optimize the low-quality areas. With a simple network structure, GL-GAN is allowed to effectively avoid the nature of imbalance by local bilevel optimization, which is accomplished by first locating low-quality areas and then optimizing them. Moreover, by using feature map cues from discriminator output, we propose the adaptive local and global optimization method(Ada-OP) for specific implementation and find that it boosts the convergence speed. Compared with the current GAN methods, our model has shown impressive performance on CelebA, CelebA-HQ and LSUN datasets.

arXiv:2007.12551

Multi-objective and categorical global optimization of photonic structures based on ResNet generative neural networks

We show that deep generative neural networks, based on global topology optimization networks (GLOnets), can be configured to perform the multi-objective and categorical global optimization of photonic devices. A residual network scheme enables GLOnets to evolve from a deep architecture, which is required to properly search the full design space early in the optimization process, to a shallow network that generates a narrow distribution of globally optimal devices. As a proof-of-concept demonstration, we adapt our method to design thin film stacks consisting of multiple material types. Benchmarks with known globally-optimized anti-reflection structures indicate that GLOnets can find the global optimum with orders of magnitude faster speeds compared to conventional algorithms. We also demonstrate the utility of our method in complex design tasks with its application to incandescent light filters. These results indicate that advanced concepts in deep learning can push the capabilities of inverse design algorithms for photonics.

arXiv:2007.12332

Image-Based Benchmarking and Visualization for Large-Scale Global Optimization

In the context of optimization, visualization techniques can be useful for understanding the behaviour of optimization algorithms and can even provide a means to facilitate human interaction with an optimizer. Towards this goal, an image-based visualization framework, without dimension reduction, that visualizes the solutions to large-scale global optimization problems as images is proposed. In the proposed framework, the pixels visualize decision variables while the entire image represents the overall solution quality. This framework affords a number of benefits over existing visualization techniques including enhanced scalability (in terms of the number of decision variables), facilitation of standard image processing techniques, providing nearly infinite benchmark cases, and explicit alignment with human perception. Furthermore, image-based visualization can be used to visualize the optimization process in real-time, thereby allowing the user to ascertain characteristics of the search process as it is progressing. To the best of the authors' knowledge, this is the first realization of a dimension-preserving, scalable visualization framework that embeds the inherent relationship between decision space and objective space. The proposed framework is utilized with 10 different mapping schemes on an image-reconstruction problem that encompass continuous, discrete, binary, combinatorial, constrained, dynamic, and multi-objective optimization. The proposed framework is then demonstrated on arbitrary benchmark problems with known optima. Experimental results elucidate the flexibility and demonstrate how valuable information about the search process can be gathered via the proposed visualization framework.

arXiv:1210.4657

Mean-Field Learning: a Survey

In this paper we study iterative procedures for stationary equilibria in games with large number of players. Most of learning algorithms for games with continuous action spaces are limited to strict contraction best reply maps in which the Banach-Picard iteration converges with geometrical convergence rate. When the best reply map is not a contraction, Ishikawa-based learning is proposed. The algorithm is shown to behave well for Lipschitz continuous and pseudo-contractive maps. However, the convergence rate is still unsatisfactory. Several acceleration techniques are presented. We explain how cognitive users can improve the convergence rate based only on few number of measurements. The methodology provides nice properties in mean field games where the payoff function depends only on own-action and the mean of the mean-field (first moment mean-field games). A learning framework that exploits the structure of such games, called, mean-field learning, is proposed. The proposed mean-field learning framework is suitable not only for games but also for non-convex global optimization problems. Then, we introduce mean-field learning without feedback and examine the convergence to equilibria in beauty contest games, which have interesting applications in financial markets. Finally, we provide a fully distributed mean-field learning and its speedup versions for satisfactory solution in wireless networks. We illustrate the convergence rate improvement with numerical examples.