Optimal Subregion Research Articles

Image Safeguarding: Reasoning with Conditional Vision Language Model and Obfuscating Unsafe Content Counterfactually

Social media platforms are being increasingly used by malicious actors to share unsafe content, such as images depicting sexual activity, cyberbullying, and self-harm. Consequently, major platforms use artificial intelligence (AI) and human moderation to obfuscate such images to make them safer. Two critical needs for obfuscating unsafe images is that an accurate rationale for obfuscating image regions must be provided, and the sensitive regions should be obfuscated (e.g. blurring) for users' safety. This process involves addressing two key problems: (1) the reason for obfuscating unsafe images demands the platform to provide an accurate rationale that must be grounded in unsafe image-specific attributes, and (2) the unsafe regions in the image must be minimally obfuscated while still depicting the safe regions. In this work, we address these key issues by first performing visual reasoning by designing a visual reasoning model (VLM) conditioned on pre-trained unsafe image classifiers to provide an accurate rationale grounded in unsafe image attributes, and then proposing a counterfactual explanation algorithm that minimally identifies and obfuscates unsafe regions for safe viewing, by first utilizing an unsafe image classifier attribution matrix to guide segmentation for a more optimal subregion segmentation followed by an informed greedy search to determine the minimum number of subregions required to modify the classifier's output based on attribution score. Extensive experiments on uncurated data from social networks emphasize the efficacy of our proposed method. We make our code available at: https://github.com/SecureAIAutonomyLab/ConditionalVLM

Discrete sizing optimization method based on dividing rectangles algorithm and local response surface for steel frame structures

It has always been the goal of structural engineers to construct safe and stable buildings using the least amount of materials. Utilizing a quick and effective way to optimize the cross-section size is crucial because conventional building design methods are impacted by the designers' knowledge, and it is challenging to prevent material waste. Due to its implicit optimization function, discrete design variables, and expensive individual evaluation, sizing optimization of high-rise buildings is very challenging to achieve. In order to effectively handle this optimization problem, a two-stage discrete sizing optimization method for high-rise buildings based on the DIviding RECTangles (DIRECT) algorithm and local response surface is proposed in this paper. The optimization method suggested in this research consists of two key stages: the global search stage and the local search stage. In the global search stage, the entire design domain is divided using a modified DIRECT algorithm to swiftly identify potentially optimal subregions that may contain the best points. In the local search stage, the local response surface model is constructed to approximate the objective and constraint functions using the sampling points from the previous stage, and the discrete optimal solution is rapidly found through mathematical iterative solving. A sizing optimization calculation program for high-rise buildings was developed in Microsoft Visual Studio 2015 on the basis of C++ to achieve automatic optimization. The new method was applied to optimize three high-rise steel frame buildings with different heights and plan shapes. The results showed that the material cost could be successfully saved compared with the conventional design, and the over-limit constraints could be adjusted automatically, which demonstrated the viability and efficacy of the two-stage optimization method.

An energy-efficient multi-objective permutation flow shop scheduling problem using an improved hybrid cuckoo search algorithm

The flow shop scheduling problem has been widely studied in recent years, but the research on multi-objective flow shop scheduling with green indicators is still relatively limited. It is urgent to strengthen the research on effective methods to solve such interesting problems. To consider the economic and environmental factors simultaneously, the paper investigates the multi-objective permutation flow shop scheduling problems (MOPFSP) which minimizes the makespan and total carbon emissions. Since MOPFSP is proved to be a NP-hard problem for more than two machines. A hybrid cuckoo search algorithm (HCSA) is proposed to solve the problems. Firstly, a largest-order-value method is proposed to enhance the performance of HCS algorithm in the solution space of MOPFSP. Then, an adaptive factor of step size is designed to control the search scopes in the evolution phases. Finally, a multi-neighborhood local search rule is addressed in order to find the optimal sub-regions obtained by the HCSA. Numerical experiments show that HCSA can solve MOPFSP efficiently.

Hypervariable Regions in 16S rRNA Genes for the Taxonomic Classification

16S ribosomal RNA (rRNA) gene sequences are reliable markers for the taxonomic classification of microbes and widely used in environmental microbiology. Production of 16S rRNA gene amplicons in large amounts, encompassing the full length of genes is not yet feasible, because of the limitations of the current sequencing techniques. They are mostly in short reads of length less than 300 base pairs. Hence, the selection of the most efficient hypervariable regions for phylogenetic analysis and taxonomic classification is a current research area. It is found that nine hypervariable regions (V1–V9), resides in bacterial 16S ribosomal RNA (rRNA) genes. Family, genus, and species-specific sequences within a given hypervariable region constitute useful targets for diagnostic assays and other scientific investigations. In this study systematic studies that compare the relative advantage of hypervariable regions grouped as V1–V2–V3, V4–V5–V6, and V7–V8–V9 for specific diagnostic goals are done. In the present research, the built in function Longest–Common–Subsequence in computer algebra package MATHEMATICA is used to create an in silico pipeline to evaluate the taxonomic classification sensitivity of the hypervariable regions compared with the corresponding full-length sequences. Conclusions: Our results suggest that V4–V5–V6 region might be an optimal sub-region for the design of universal primers with superior phylogenetic resolution for bacterial phyla.

Continuum approximation models for joint delivery systems using trucks and drones

ABSTRACT This research focuses on exploring the economics related to the joint delivery system using trucks and drones. Transportation distances and costs are approximated as simple functions using continuum approximation (CA) methods, which can keep key issues and tradeoffs in focus. The main contribution is to develop cost models using the derived methods and gain a greater understanding of delivery activities by focusing on the tradeoffs between major components, which are valuable for decision-makers: choose economical delivery mode based on customer density, partition service region into optimal sub-regions, and obtain an optimal delivery ratio between trucks and drones.

Sensitivity and correlation of hypervariable regions in 16S rRNA genes in phylogenetic analysis.

BackgroundProkaryotic 16S ribosomal RNA (rRNA) sequences are widely used in environmental microbiology and molecular evolution as reliable markers for the taxonomic classification and phylogenetic analysis of microbes. Restricted by current sequencing techniques, the massive sequencing of 16S rRNA gene amplicons encompassing the full length of genes is not yet feasible. Thus, the selection of the most efficient hypervariable regions for phylogenetic analysis and taxonomic classification is still debated. In the present study, several bioinformatics tools were integrated to build an in silico pipeline to evaluate the phylogenetic sensitivity of the hypervariable regions compared with the corresponding full-length sequences.ResultsThe correlation of seven sub-regions was inferred from the geodesic distance, a parameter that is applied to quantitatively compare the topology of different phylogenetic trees constructed using the sequences from different sub-regions. The relationship between different sub-regions based on the geodesic distance indicated that V4-V6 were the most reliable regions for representing the full-length 16S rRNA sequences in the phylogenetic analysis of most bacterial phyla, while V2 and V8 were the least reliable regions.ConclusionsOur results suggest that V4-V6 might be optimal sub-regions for the design of universal primers with superior phylogenetic resolution for bacterial phyla. A potential relationship between function and the evolution of 16S rRNA is also discussed.Electronic supplementary materialThe online version of this article (doi:10.1186/s12859-016-0992-y) contains supplementary material, which is available to authorized users.

Shadow Prices in Territory Division

We consider a geographic optimization problem in which we are given a region R, a probability density function f(⋅) defined on R, and a collection of n utility density functions u i (⋅) defined on R. Our objective is to divide R into n sub-regions R i so as to “balance” the overall utilities on the regions, which are given by the integrals $\iint _{R_{i}}f(x)u_{i}(x)\, dA$ . Using a simple complementary slackness argument, we show that (depending on what we mean precisely by “balancing” the utility functions) the boundary curves between optimal sub-regions are level curves of either the difference function u i (x) − u j (x) or the ratio u i (x)/u j (x). This allows us to solve the problem of optimally partitioning the region efficiently by reducing it to a low-dimensional convex optimization problem. This result generalizes, and gives very short and constructive proofs of, several existing results in the literature on equitable partitioning for particular forms of f(⋅) and u i (⋅). We next give two economic applications of our results in which we show how to compute a market-clearing price vector in an aggregate demand system or a variation of the classical Fisher exchange market. Finally, we consider a dynamic problem in which the density function f(⋅) varies over time (simulating population migration or transport of a resource, for example) and derive a set of partial differential equations that describe the evolution of the optimal sub-regions over time. Numerical simulations for both static and dynamic problems confirm that such partitioning problems become tractable when using our methods.

Centroidal Area-Constrained Partitioning for Robotic Networks

We consider the problem of optimal coverage with area-constraints in a mobile multi-agent system. For a planar environment with an associated density function, this problem is equivalent to dividing the environment into optimal subregions such that each agent is responsible for the coverage of its own region. In this paper, we design a continuous-time distributed policy which allows a team of agents to achieve a convex area-constrained partition of a convex workspace. Our work is related to the classic Lloyd algorithm, and makes use of generalized Voronoi diagrams. We also discuss practical implementation for real mobile networks. Simulation methods are presented and discussed.

Simplicial Lipschitz optimization without the Lipschitz constant

In this paper we propose a new simplicial partition-based deterministic algorithm for global optimization of Lipschitz-continuous functions without requiring any knowledge of the Lipschitz constant. Our algorithm is motivated by the well-known Direct algorithm which evaluates the objective function on a set of points that tries to cover the most promising subregions of the feasible region. Almost all previous modifications of Direct algorithm use hyper-rectangular partitions. However, other types of partitions may be more suitable for some optimization problems. Simplicial partitions may be preferable when the initial feasible region is either already a simplex or may be covered by one or a manageable number of simplices. Therefore in this paper we propose and investigate simplicial versions of the partition-based algorithm. In the case of simplicial partitions, definition of potentially optimal subregion cannot be the same as in the rectangular version. In this paper we propose and investigate two definitions of potentially optimal simplices: one involves function values at the vertices of the simplex and another uses function value at the centroid of the simplex. We use experimental investigation to compare performance of the algorithms with different definitions of potentially optimal partitions. The experimental investigation shows, that proposed simplicial algorithm gives very competitive results to Direct algorithm using standard test problems and performs particularly well when the search space and the numbers of local and global optimizers may be reduced by taking into account symmetries of the objective function.

Mixture-process variable approach to optimize a microemulsion electrokinetic chromatography method for the quality control of a nutraceutical based on coenzyme Q10

In recent years, multivariate optimization has played an increasing role in analytical method development. ICH guidelines recommend using statistical design of experiments to identify the design space, in which multivariate combinations of composition variables and process variables have been demonstrated to provide quality results. Considering a microemulsion electrokinetic chromatography method (MEEKC), the performance of the electrophoretic run depends on the proportions of mixture components (MCs) of the microemulsion and on the values of process variables (PVs). In the present work, for the first time in the literature, a mixture-process variable (MPV) approach was applied to optimize a MEEKC method for the analysis of coenzyme Q10 (Q10), ascorbic acid (AA), and folic acid (FA) contained in nutraceuticals. The MCs (buffer, surfactant–cosurfactant, oil) and the PVs (voltage, buffer concentration, buffer pH) were simultaneously changed according to a MPV experimental design. A 62-run MPV design was generated using the I-optimality criterion, assuming a 46-term MPV model allowing for special-cubic blending of the MCs, quadratic effects of the PVs, and some MC-PV interactions. The obtained data were used to develop MPV models that express the performance of an electrophoretic run (measured as peak efficiencies of Q10, AA, and FA) in terms of the MCs and PVs. Contour and perturbation plots were drawn for each of the responses. Finally, the MPV models and criteria for the peak efficiencies were used to develop the design space and an optimal subregion (i.e., the settings of the mixture MCs and PVs that satisfy the respective criteria), as well as a unique optimal combination of MCs and PVs.

Global Optimization of Combined Region Aggregation and Leveling Model

The advent and widespread use of innovative spatial analysis technologies, such as geographic information systems (GIS), computer aided design (CAD) systems, and global positioning systems (GPS), have prompted great interest in spatial optimization. The tasks of selecting an optimal subregion from a larger region—region aggregation—and determining an optimal strategy for cutting and filling that subregion to a uniform elevation—land leveling—are examples of spatial analyses that can benefit from these powerful computer technologies. The combined region aggregation and leveling problem is a complex spatial problem that often involves the comprehensive consideration of multiple, incommensurate, and often conflicting objectives, while at the same time satisfying a set of prespecified physical and logical constraints. Traditionally, these two problems are solved separately, often precluding the identification of global optima. Through this research, a multiobjective integer programming model that considers these problems simultaneously is formulated, a computational algorithm for solving the model is presented, and numerical results that demonstrate the efficiency and effectiveness of this procedure are discussed. Computational experiments report polynomial complexity of the heuristic procedure against exponential worst-case complexity of traditional enumerative methods.