Binary Constraints Research Articles

Hierarchical Multi-label Classification using Fully Associative Ensemble Learning

Traditional flat classification methods (e.g., binary or multi-class classification) neglect the structural information between different classes. In contrast, Hierarchical Multi-label Classification (HMC) considers the structural information embedded in the class hierarchy, and uses it to improve classification performance. In this paper, we propose a local hierarchical ensemble framework for HMC, Fully Associative Ensemble Learning (FAEL). We model the relationship between each class node’s global prediction and the local predictions of all the class nodes as a multi-variable regression problem with Frobenius norm or l1 norm regularization. It can be extended using the kernel trick, which explores the complex correlation between global and local prediction. In addition, we introduce a binary constraint model to restrict the optimal weight matrix learning. The proposed models have been applied to image annotation and gene function prediction datasets with tree structured class hierarchy and large scale visual recognition dataset with Direct Acyclic Graph (DAG) structured class hierarchy. The experimental results indicate that our models achieve better performance when compared with other baseline methods.

Nonlinear Discrete Cross-Modal Hashing for Visual-Textual Data

Hashing techniques have been widely adopted for cross-modal retrieval due to their low storage cost and fast query speed. Recently, some unimodal hashing methods have tried to directly optimize the objective function with discrete binary constraints. Inspired by these methods, the authors propose a novel supervised cross-modal hashing method called Discrete Cross-Modal Hashing (DCMH) to learn the binary codes without relaxing them. DCMH is formulated through semantic similarity reconstruction, and it learns binary codes for use as ideal features for classification. Furthermore, DCMH alternately updates binary codes for each modality, and its discrete hashing codes are learned efficiently, bit by bit, which is quite promising for large-scale datasets. To evaluate the effectiveness of the proposed discrete optimization, the authors optimize their objective function in a relax-and-threshold manner. Extensive empirical results on both image-text and image-tag datasets demonstrate that DCMH is a significant improvement over previous approaches in terms of training time and retrieval performance.

Identity authentication based on two-beam interference and nonlinear correlation

In this paper, a new approach to identity authentication is proposed, which takes advantage of the two-beam interference setup and the nonlinear correlation technique. According to the traditional two-beam interference encryption/decryption structure, we design a modified iterative phase retrieval algorithm (MIPRA), which takes the random binary amplitudes as the constraints at the input plane to encode different images (standard reference images) into a set of sparse phase distributions. In the MIPRA, a given random phase distribution serves as a system lock, and it is placed at one of the arms of the two-beam interference setup and keeps unchanged in the whole iterative phase retrieval algorithm but equivalently provides a fixed shifting vector toward the output complex amplitude field. While the peak-to-correlation value (between the output intensity and the original image) reaches a presetting threshold value, or the iterative numer of time reaches a presetting maximum value, the MIPRA stops. Here, the phase lock is assumed to be the same for all the users and thus it is placed and fixed in the system, while the calculated phase distributions vary from the MIPRA to different binary constraints, which are related to different users. Meanwhile, we also study an extension version of the proposed method. By using a superposition multiplexing technique and a nonlinear correlation technique, we can realize a function of hierarchical authentication for various kinds of users through a similar but more smart decision strategy. For example, we adopt the MIPRA four times with different constraints (random binary amplitude distribution) to obtain four phase distributions, the sum of them will be regarded as a final phase key and is designed to the user with the highest privilege. He is then able to pass all the authentication process for each standard reference image with his multiplexed phase key, that is to say, there are obvious peaks in all the nonlinear correlation maps between all the output images and the corresponding standard reference images. In a similar way, the user with the lowest privilege can only pass one authentication process. Compared with the previous identity authentication methods in the optical security area, the phase key for each user, no matter what level he belongs to, is easy to be stored and transmitted because its distinguishing feature of sparsity. It is worthwhile to note that the cross-talk between different output images are very low and will has no effect on the authentication decision since we deliberately assemble all the binary distributions, which act as constraints at the input plane in the MIPRA. Moreover, the output results are all noise-like distributions, which makes it nearly impossible for any potential intruders to find any clues of the original standard reference images. However, on the other hand, with the nonlinear correlation technique, we can easily extract enough information from these noise-like output results to authorize any users, usually we can obtain an obvious peak at the center of the correlation results but there is no peak if we adopt the traditional correlation algorithms. This feature helps reduce the risk of information leakage, thereby providing an additional protection layer. Also, weinvestigate the robustness properties by taking the sparsity ratio, Gaussian noise, and shear/occluded attack into consideration. Some previous tests alsoindicated that our scheme can resist the attack employing incorrect random phase keys. Theoretical analysis and a series simulation results are provided to verify the feasibility and effectiveness of the proposed scheme.

An optimization modeling of coordinated traffic signal control based on the variational theory and its stochastic extension

Abstract: This study considers an optimal coordinated traffic signal control under both deterministic and stochastic demands. We first present a new mixed integer linear programming (MILP) for the deterministic signal optimization wherein traffic flow is modeled based on the variational theory and the constraints on a signal control pattern are linearly formulated. The resulting MILP has a clear network structure and requires fewer binary variables and constraints as compared with those in the existing formulations. We then extend the problem so as to treat the stochastic fluctuations in traffic demand. We here develop an accurate and efficient approximation method of expected delays and a solution method for the stochastic version of the signal optimization by exploiting the network structure of the problem. Using a set of proposed methods, we finally examine the optimal control parameters for deterministic and stochastic coordinated signal controls and discuss their characteristics.

On Downlink Resource Allocation for SWIPT in Small Cells in a Two-Tier HetNet

We study the downlink resource allocation problem for simultaneous wireless information and power transfer (SWIPT) in small cells underlaying a macrocell in a two-tier heterogeneous network. For SWIPT, we consider both time-switching and power-splitting approaches. We determine downlink transmit power of small cell base stations along with time-switching/power-splitting variables for SWIPT to jointly optimize energy harvesting rate and achievable throughput of small cell users while ensuring minimum throughput of macrocell user. In the time-switching approach, the resource allocation problem is formulated as a mixed-integer non-linear programming (MINLP) problem. The formulated MINLP problem is solved by relaxing the binary integer constraint and then identifying the condition at which the obtained solution satisfies that constraint. In both the time-switching and power-splitting approaches, in the presence of non-negligible co-tier interference, the formulated problem is solved sub-optimally by iteratively maximizing the minorant of the non-convex objective function. A special case with negligible co-tier interference is considered in the time-switching approach and the optimal solution is obtained by using convex optimization techniques. Numerical results demonstrate significant gain in energy harvesting rate when macrocell users have flexible interference tolerance levels in time-switching approach, highlight the improvement in energy harvesting rate in the presence of co-tier interference signal, and reveal interesting trade-off in achievable throughput and energy harvesting rate.

Safe binary particle swam algorithm for an enhanced unsupervised label refinement in automatic face annotation

Mining web facial images on the internet has become as a profitable and important paradigm towards auto face annotation technique. The unsupervised label refinement (ULR) is an effective method that can fix weakly labeled facial images data which are collected from the internet and included some images with wrong label. In order to improve the correction accuracy of ULR, particle swarm optimization (PSO) and binary particle swarm optimization (BPSO) are used for solving binary constraint optimization task in this study. A novel method named safe binary particle swam optimization (SBPSO) is also proposed to improve BPSO which has the probability over range problem for using the ULR. In addition, SBPSO is also employed for an enhanced ULR (EULR) objective function which is created by modifying the original formula of ULR to improve the accuracy of labeled facial image. An experimental database is queried from IMDb website which collected the actors who were bored in 1950 to 1990. Some error flags are randomly added in the database for the correction tests by different methods. The results showed that the SBPSO Algorithm for the EULR in automatic face annotation have the better label correction rate and convergence effect.

Identification of unknown parameters of a single diode photovoltaic model using particle swarm optimization with binary constraints

Abstract-Photo-voltaic (PV) is a static medium to convert solar energy directly into electricity. In order to predict the performance of a PV system before being installed, a reliable and accurate model design of PV systems is essential. To validate the design of a PV system like maximum power point (MPP) and micro-grid system through simulation, an accurate solar PV model is required. However, information provided by manufacturers in data sheets is not sufficient for simulating the characteristic of a PV module under normal as well as under diverse environmental conditions. In this paper, a particle swarm optimization (PSO) technique with binary constraints has been presented to identify the unknown parameters of a single diode model of solar PV module. Multi-crystalline and mono-crystalline technologies based PV modules are considered under the present study. Based on the results obtained, it has been found that PSO algorithm yields a high value of accuracy irrespective of temperature variations.

Revisiting restricted path consistency

Restricted path consistency (RPC) is a strong local consistency for binary constraints that was proposed 20 years ago and was identified as a promising alternative to arc consistency (AC) in an early experimental study of local consistencies for binary constraints. However, in contrast to other strong local consistencies such as singleton arc consistency (SAC) and max restricted path consistency (maxRPC), it has been neglected since then. In this paper we revisit RPC. First we propose RPC3, a new RPC algorithm that is very easy to implement and can be efficiently applied throughout search. Then we perform a wide experimental study of RPC3 and a light version that achieves an approximation of RPC, comparing them to state-of-the-art AC and maxRPC algorithms. Experimental results obtained under various solver settings, regarding the branching scheme, the variable ordering heuristic, and restarts, clearly show that light RPC is by far more efficient than both AC and maxRPC when applied throughout search. We also examine the experimental behaviour of an algorithm for a simple extension of RCP called 2-RPC, showing that it is competitive with RPC3, and better than both AC and maxRPC. Overall, our results strongly suggest that it is time to reconsider the established perception that MAC is the best general purpose method for solving binary CSPs.

Solar Farm Suitability Using Geographic Information System Fuzzy Sets and Analytic Hierarchy Processes: Case Study of Ulleung Island, Korea

Solar farm suitability in remote areas will involve a multi-criteria evaluation (MCE) process, particularly well suited for the geographic information system (GIS) environment. Photovoltaic (PV) solar farm criteria were evaluated for an island-based case region having complex topographic and regulatory criteria, along with high demand for low-carbon local electricity production: Ulleung Island, Korea. Constraint variables that identified areas forbidden to PV farm development were consolidated into a single binary constraint layer (e.g., environmental regulation, ecological protection, future land use). Six factor variables were selected as influential on-site suitability within the geospatial database to seek out increased annual average power performance and reduced potential investment costs, forming new criteria layers for site suitability: solar irradiation, sunshine hours, average temperature in summer, proximity to transmission line, proximity to roads, and slope. Each factor variable was normalized via a fuzzy membership function (FMF) and parameter setting based on the local characteristics and criteria for a fixed axis PV system. Representative weighting of the relative importance for each factor variable was assigned via pairwise comparison completed by experts. A suitability index (SI) with six factor variables was derived using a weighted fuzzy summation method. Sensitivity analysis was conducted to assess four different SI based on the development scenarios (i.e., the combination of factors being considered). From the resulting map, three highly suitable regions were suggested and validated by comparison with satellite images to confirm the candidate sites for solar farm development. The GIS-MCE method proposed can also be applicable widely to other PV solar farm site selection projects with appropriate adaption for local variables.

Multiplexed Spectral Imaging of 120 Different Fluorescent Labels.

The number of fluorescent labels that can unambiguously be distinguished in a single image when acquired through band pass filters is severely limited by the spectral overlap of available fluorophores. The recent development of spectral microscopy and the application of linear unmixing algorithms to spectrally recorded image data have allowed simultaneous imaging of fluorophores with highly overlapping spectra. However, the number of distinguishable fluorophores is still limited by the unavoidable decrease in signal to noise ratio when fluorescence signals are fractionated over multiple wavelength bins. Here we present a spectral image analysis algorithm to greatly expand the number of distinguishable objects labeled with binary combinations of fluorophores. Our algorithm utilizes a priori knowledge about labeled specimens and imposes a binary label constraint on the unmixing solution. We have applied our labeling and analysis strategy to identify microbes labeled by fluorescence in situ hybridization and here demonstrate the ability to distinguish 120 differently labeled microbes in a single image.

Efficient local search for L_1 and L_2 binary matrix factorization

Rank K Binary Matrix Factorization (BMF) approximates a binary matrix by the product of two binary matrices of lower rank, K. Several researchers have addressed this problem, focusing on either approximations of rank 1 or higher, using either the L 1 or L 2-norms for measuring the quality of the approximation. The rank 1 problem (for which the L 1 and L 2-norms are equivalent) has been shown to be related to the Integer Linear Programming (ILP) problem. We first show here that the alternating strategy with the L 2-norm, at the core of several methods used to solve BMF, can be reformulated as an Unconstrained Binary Quadratic Programming (UBQP) problem. This reformulation allows us to use local search procedures designed for UBQP in order to improve the solutions of BMF. We then introduce a new local search dedicated to the BMF problem. We show in particular that this solution is in average faster than the previously proposed ones. We then assess its behavior on several collections and methods and show that it significantly improves methods targeting the L 2-norms on all the datasets considered; for the L 1-norm, the improvement is also significant for real, structured datasets and for the BMF problem without the binary reconstruction constraint.

Multi-View 3D Human Tracking in Crowded Scenes

This paper presents a robust multi-view method for tracking people in 3D scene. Our method distinguishes itself from previous works in two aspects. Firstly, we define a set of binary spatial relationships for individual subjects or pairs of subjects that appear at the same time, e.g. being left or right, being closer or further to the camera, etc. These binary relationships directly reflect relative positions of subjects in 3D scene and thus should be persisted during inference. Secondly, we introduce an unified probabilistic framework to exploit binary spatial constraints for simultaneous 3D localization and cross-view human tracking. We develop a cluster Markov Chain Monte Carlo method to search the optimal solution. We evaluate our method on both public video benchmarks and newly built multi-view video dataset. Results with comparisons showed that our method could achieve state-of-the-art tracking results and meter-level 3D localization on challenging videos.

Exploiting Binary Floating-Point Representations for Constraint Propagation

Floating-point computations are quickly finding their way in the design of safety- and mission-critical systems, despite the fact that designing floating-point algorithms is significantly more difficult than designing integer algorithms. For this reason, verification and validation of floating-point computations are hot research topics. An important verification technique, especially in some industrial sectors, is testing. However, generating test data for floating-point intensive programs proved to be a challenging problem. Existing approaches usually resort to random or search-based test data generation, but without symbolic reasoning it is almost impossible to generate test inputs that execute complex paths controlled by floating-point computations. Moreover, because constraint solvers over the reals or the rationals do not natively support the handling of rounding errors, the need arises for efficient constraint solvers over floating-point domains. In this paper, we present and fully justify improved algorithms for the propagation of arithmetic IEEE 754 binary floating-point constraints. The key point of these algorithms is a generalization of an idea by B. Marre and C. Michel that exploits a property of the representation of floating-point numbers.

An Improved Energy-Awareness Formulation for General Precedence Continuous-Time Scheduling Models

Efficient frameworks for scheduling of energy-intensive industries are important enablers for the industrial implementation of the demand-side management concept (DSM). DSM enables enterprises to cut their energy cost as well as to support the grid in dealing with fluctuating availability of energy. In this work, we present an improvement of the energy-aware production scheduling strategy reported in Hadera et al.,1 which is based upon a continuous-time scheduling formulation. The improvement concerns the modeling of the coupling between the scheduling problem and the computation of the energy bill. The resulting optimization problem is solved by the heuristic bilevel decomposition from Hadera et al.2 Numerical studies showed that although the proposed formulation and heuristic decomposition does not guarantee to obtain the global minimum it does provide good quality solutions within reasonable solution times. The new energy-awareness formulation reduces the number of binary variables and constraints and ...

Challenging the presence of scalar charge and dipolar radiation in binary pulsars

Corrections to general relativity that introduce long-ranged scalar fields which are non-minimally coupled to curvature typically predict that neutron stars possess a non-trivial scalar field profile. An observer far from a star is most sensitive to the spherically-symmetric piece of this profile that decays linearly with the inverse of the distance, the so-called scalar charge, which is related to the emission of dipolar radiation from compact binaries. The presence of dipolar radiation has the potential to very strongly constrain extended theories of gravity. These facts may lead people to believe that gravitational theories with long-ranged scalar fields have already been constrained strongly from binary pulsar observations. Here we challenge this "lore" by investigating the decoupling limit of Gauss-Bonnet gravity as an example, in which the scalar field couples linearly to the Gauss-Bonnet density in the action. We prove a theorem that neutron stars in this theory cannot possess a scalar charge. Thus Gauss-Bonnet gravity evades the strong binary pulsar constraints on dipole radiation. We discuss the astrophysical systems yielding the best constraints on Gauss-Bonnet gravity and related quadratic gravity theories. To achieve this we explicitly compute the scalar charge for slowly-rotating neutron stars in quadratic gravity theories. In generic case, either neutron star-binary or neutron star-black hole systems can be used to constrain the theory, but Gauss-Bonnet gravity evades the neutron star-binary constraints. However, black holes in Gauss-Bonnet gravity do anchor scalar charge. The best constraints on Gauss-Bonnet gravity will thus come from black hole observations, for example via pulsar-black hole binaries. We estimate these constraints to be ten times better than the current estimated bound, and also include estimated constraints on generic quadratic gravity theories.

Review of binary tournament constraint handling technique in NSGA II for optimal control of combined sewer systems

Binary tournament constraint handling technique in Non Sorted Genetic Algorithm II (NSGA II) has both advantages and disadvantages. This constraint handling technique is fairly simple; however, it solves the issues in identifying the penalty functions for multi- objective optimization problems. This paper reviews the constraint handling approach in NSGA II and compares it with an alternative constraint handling approach for optimal control of combined sewer systems. Interesting results are presented and it can be concluded that the alternative constraint handling approach produces better results; however, this alternative constraint handling approach has to be used with adequate care.

Mixed Mode Transmission and Resource Allocation for D2D Communication

In cellular communication systems with optional device-to-device (D2D) links, user equipments (UEs) can operate in either D2D mode or cellular mode for data transport. This work introduces mixed-mode D2D communication in which D2D links can operate in multiple modes through resource multiplexing. Within this framework, we study the problem of maximizing weighted D2D sum rate under cellular rate constraints by optimizing mixed-mode allocation and resource allocation in term of transmit power and subchannel assignment. Due to nonconvex cellular rate constraints and binary constraints of subchannel allocation, this problem is a nonconvex mixed-integer problem that is generally difficult to solve. We propose a two-step approach by introducing energy-splitting variables such that mixed-mode allocation and resource allocation can be decoupled and optimized independently. The resulting algorithm can be distributive, requires little signaling overhead, and has low computational complexity. We present extensive numerical results to demonstrate the practicality of our proposed algorithm with regard to various network parameters.

Switch programming of reflectivity control devices for the coupled dynamics of a solar sail

As demonstrated in the Interplanetary Kite-craft Accelerated by Radiation Of the Sun (IKAROS), reflectivity control devices (RCDs) are switched on or off independently with each other, which has nevertheless been ignored by many previous researches. This paper emphasizes the discrete property of RCDs, and aims to obtain an appropriate switch law of RCDs for a rigid spinning solar sail. First, the coupled attitude–orbit dynamics is derived from the basic solar force and torque model into an underdetermined linear system with a binary set constraint. Subsequently, the coupled dynamics is reformulated into a constrained quadratic programming and a basic gradient projection method is designed to search for the optimal solution. Finally, a circular sail flying in the Venus rendezvous mission demonstrates the model and method numerically, which illustrates approximately 103km terminal position error and 0.5m/s terminal velocity error as 80 independent RCDs are switched on or off appropriately.

Predictive path following with arrival time awareness for waterborne AGVs

Large ports are seeking innovative logistical ways to improve their competitiveness world-wide. This article proposes waterborne AGVs, inspired by conventional automated guided vehicles and autonomous surface vessels, for transport over water. A predictive path following with arrival time awareness controller is proposed for such waterborne AGVs. The controller is able to achieve smooth tracking and energy efficiency with arrival time awareness for transport oriented applications. Tracking errors are conveniently formulated with vessel dynamics modeled in connected reference path coordinate systems and a coordinate transformation at switching coordinate systems. Binary decision variables and logic constraints based on an along-track state are proposed for modeling switches in the framework of Model Predictive Control (MPC) so that overshoots are avoided. Moreover, timing-aware along-track references are generated by a two-level double integrator scheme. The lower level is embedded in online MPC optimizations for smooth tracking. The higher level solves a mixed-integer quadratic programming problem considering distance-to-go and time-to-go before each MPC optimization. References over the next prediction horizon are generated being aware of the requirements on arrival time. Furthermore, successive linearizations of nonlinear vessel dynamics about a shifted previous optimal system trajectory are implemented to maintain a trade-off between computational complexity and optimality. Simulation results of two industrially relevant Inter Terminal Transport case studies illustrate the effectiveness of the proposed modeling and control design for waterborne AGVs.

Improving degree-based variable ordering heuristics for solving constraint satisfaction problems

In this paper, we improved two classical degree-based variable ordering heuristics, $$\frac{\textit{Dom}}{\textit{Ddeg}}$$DomDdeg and $$\frac{\textit{Dom}}{\textit{Wdeg}}$$DomWdeg. We propose a method using the summation of constraint tightness in degree-based heuristics. We also propose two methods to calculate dynamic constraint tightness for binary extensional constraints and non-binary intensional constraints respectively. Our work shows how constraint tightness can be practically used to guide search. We performed a number of experiments on some benchmark instances. The results have shown that, the new heuristics improve the classical ones by both computational time and search tree nodes and they are more efficient than some other successful heuristics on the instances where the classical heuristics work well.