Diversity Tradeoff Research Articles

Addressing the so-called validity–diversity trade-off: Exploring the practicalities and legal defensibility of Pareto-optimization for reducing adverse impact within personnel selection

AbstractIt is necessary for personnel selection systems to be effective, fair, and legally appropriate. Sometimes these goals are complementary, whereas other times they conflict (leading to the so-called “validity-diversity dilemma”). In this practice forum, we trace the history and legality of proposed approaches for simultaneously maximizing job performance and diversity through personnel selection, leading to a review of a more recent method, the Pareto-optimization approach. We first describe the method at various levels of complexity and provide guidance (with examples) for implementing the technique in practice. Then, we review the potential points at which the method might be challenged legally and present defenses against those challenges. Finally, we conclude with practical tips for implementing Pareto-optimization within personnel selection.

Structured Light in Turbulence

Optical communication is an integral part of the modern economy, having all but replaced electronic communication systems. Future growth in bandwidth appears to be on the horizon using structured light, encoding information into the spatial modes of light, and transmitting them down fibre and free-space, the latter crucial for addressing last mile and digitally disconnected communities. Unfortunately, patterns of light are easily distorted, and in the case of free-space optical communication, turbulence is a significant barrier. Here we review recent progress in structured light in turbulence, first with a tutorial style summary of the core concepts, before highlighting the present state-of-the-art in the field. We support our review with new experimental studies that reveal which types of structured light are best in turbulence, the behaviour of vector versus scalar light in turbulence, the trade-off of diversity and multiplexing, and how turbulence models can be exploited for enhanced optical signal processing protocols. This comprehensive treatise will be invaluable to the large communities interested in free-space optical communication with spatial modes of light.

A modified particle swarm optimization based on decomposition with different ideal points for many-objective optimization problems

Many evolutionary algorithms have been proposed for multi-/many-objective optimization problems; however, the tradeoff of convergence and diversity is still the challenge for optimization algorithms. In this paper, we propose a modified particle swarm optimization based on decomposition framework with different ideal points on each reference vector, called MPSO/DD, for many-objective optimization problems. In the MPSO/DD algorithm, the decomposition strategy is used to ensure the diversity of the population, and the ideal point on each reference vector can draw the population converge faster to the optimal front. The position of each individual will be updated by learning the demonstrators in its neighborhood that have less distance to the ideal point along the reference vector. Eight state-of-the-art evolutionary multi-/many-objective optimization algorithms are adopted to compare the performance with MPSO/DD for solving many-objective optimization problems. The experimental results on seven DTLZ test problems with 3, 5, 8, 10, 15 and 20 objectives, respectively, show the efficiency of our proposed method on solving problems with high-dimensional objective space.

Efficient network architecture search via multiobjective particle swarm optimization based on decomposition

The efforts devoted to manually increasing the width and depth of convolutional neural network (CNN) usually require a large amount of time and expertise. It has stimulated a rising demand of neural architecture search (NAS) over these years. However, most popular NAS approaches solely optimize for low prediction error without penalizing high structure complexity. To this end, this paper proposes MOPSO/D-Net, a CNN architecture search method with multiobjective particle swarm optimization based on decomposition (MOPSO/D). The main goal is to reformulate NAS as a multiobjective evolutionary optimization problem, where the optimal architecture is learned by minimizing two conflicting objectives, namely the error rate of classification and number of parameters of the network. Along with the hybrid binary encoding and adaptive penalty-based boundary intersection, an improved MOPSO/D is further proposed to solve the formulated multiobjective NAS and provide diverse tradeoff solutions. Experimental studies verify the effectiveness of MOPSO/D-Net compared with current manual and automated CNN generation methods. The proposed algorithm achieves impressive classification performance with a small number of parameters on each of two benchmark datasets, particularly, 0.4% error rate with 0.16M params on MNIST and 5.88% error rate with 8.1M params on CIFAR-10, respectively.

Multiple Active Spatial Modulation: A Possibility of More Than Spatial Multiplexing

For multiple active spatial modulation (MASM) technique, this letter formulates the minimum number of transmit antennas $(N_{\mathrm {t}})$ and the required number of active RF chains ( $N_{\mathrm {a}}$ ) to support the following two statements. First, it is proven that MASM can achieve similar spectral efficiency to conventional spatial multiplexing (SMx) with smaller $N_{\mathrm {t}}$ and $N_{\mathrm {a}}$ , and less receive antennas ( $N_{\mathrm {r}}$ ). Secondly, for operation with the same $N_{\mathrm {t}}$ in both techniques, MASM is shown to achieve higher spectral efficiency while using fewer $N_{\mathrm {a}}$ and $N_{\mathrm {r}}$ . Furthermore, the low spectral efficiency and diversity tradeoff in MASM is highlighted, and a proposal of an enhanced MASM (E-MASM) technique is shown to achieve a better tradeoff between spectral efficiency and diversity as compared to SMx. In comparison to SMx and some prominent space modulation techniques, E-MASM is shown to achieve higher reliability, while MASM achieves superior spectral efficiency.

Analysing exposure diversity in collaborative recommender systems—Entropy fusion approach

Recommender Systems are considered as essential business tools to leverage the potential growth of on-line services. Neighbourhood based collaborative filtering, a successful recommendation approach has mainly focused on improving accuracy of predictions. From user point of view, it is more valuable to obtain novel and diverse recommendations rather than monotonic preferences. Ratings given by a user for different categories of items are considered as a tool to access user exposure diversity which signifies his creative and divergent thinking. On the other hand, pair of items is concordant if highly correlated users agree in rating the items. Based on the user exposure diversity and item concordance, the neighbourhood selection process of item based collaborative recommender systems is refined. Rating predictions are made based on the newly selected neighbours. The performance of the proposed approach is investigated for accuracy and diversity of predictions on Movielens data sets. The results demonstrate that the proposed approach outperforms the state of the art recommendation approaches which address accuracy–diversity trade off. Statistical analysis is done to prove the efficiency of the proposed approach.

A Pragmatic Testing-Eligibility Framework for Population Mutation Screening: The Example of BRCA1/2.

Eligibility guidelines for genetic testing may be revisited, given technological advances, plummeting costs, and proposals for population mutation screening. A key property of eligibility criteria is the tradeoff between the number of mutation carriers identified versus population members tested. We assess the fractions of mutation carriers identified, versus women undergoing mutation testing, for BRCA1/2 founder mutation screening in U.S. Ashkenazi-Jewish women. BRCA1/2 carrier probabilities, based on personal/family history, were calculated using the risk-prediction tool BRCAPRO for 4,589 volunteers (102 mutation carriers) in the population-based Washington Ashkenazi Study. For each carrier-probability threshold between 0% and 10%, we compared the percentage of founder mutations detected versus the percentage of women requiring mutation testing. PCR mutation testing was conducted at the NIH for the 3 Ashkenazi-Jewish founder mutations (5382insC and 185delAG in BRCA1, and 6174delT in BRCA2). Identifying 90% of BRCA1/2 founder mutations required testing the 60% of Ashkenazi-Jewish women with carrier probabilities exceeding 0.56%, potentially avoiding mutation testing for approximately 0.7 to 1.1 million U.S. Ashkenazi-Jewish women. Alternatively, testing the 44% whose carrier probability exceeded 0.78% identified 80% of mutation carriers, increasing to 89% of mutation carriers when accounting for cascade testing triggered after mutation-positive daughters were identified by screening. We present data on all carrier-probability thresholds, e.g., a 5% threshold identified 46% of mutation carriers while testing 10% of women. Different carrier-probability thresholds offered diverse tradeoffs between numbers of identified mutation carriers versus women tested. Low carrier-probability thresholds identified 90% of BRCA1/2 founder mutation carriers, without testing approximately 1 million U.S. Ashkenazi-Jewish women with lowest carrier probabilities. In general, this risk-based framework could provide useful new options to consider during eligibility-criteria development for population mutation screening.

Successive Amplify-and-Forward Relaying With Network Interference Cancellation

Successive relaying holds the promise of recovering the multiplexing loss due to the half-duplex constraint, by allowing the source and relays to transmit messages simultaneously, but it may cause severe inter-relay interference (IRI). To overcome this, we apply an amplify-and-cancel method proposed in our previous work by Chen et al. into the successive amplify-and-forward (AF) relay model, thereby conceiving successive AF relaying with network interference cancellation. The proposed protocol benefits from low complexity because IRI is mitigated by linear analog processing without decoding any inter-relay signals. More specifically, a relay keeps receiving signals from the source and other relays before its own transmission, in order to obtain the prior knowledge of IRI that can be used to mitigate the IRI. We present an iterative expression of the power of the residual interference, based on which the achievable rate and the outage probability of the proposed scheme are derived. We further minimize its outage probability by optimizing the order of relays and present the optimal diversity and multiplexing tradeoff.

Multi-Timescale Drowsiness Characterization Based on a Video of a Driver's Face.

Drowsiness is a major cause of fatal accidents, in particular in transportation. It is therefore crucial to develop automatic, real-time drowsiness characterization systems designed to issue accurate and timely warnings of drowsiness to the driver. In practice, the least intrusive, physiology-based approach is to remotely monitor, via cameras, facial expressions indicative of drowsiness such as slow and long eye closures. Since the system’s decisions are based upon facial expressions in a given time window, there exists a trade-off between accuracy (best achieved with long windows, i.e., at long timescales) and responsiveness (best achieved with short windows, i.e., at short timescales). To deal with this trade-off, we develop a multi-timescale drowsiness characterization system composed of four binary drowsiness classifiers operating at four distinct timescales (5 s, 15 s, 30 s, and 60 s) and trained jointly. We introduce a multi-timescale ground truth of drowsiness, based on the reaction times (RTs) performed during standard Psychomotor Vigilance Tasks (PVTs), that strategically enables our system to characterize drowsiness with diverse trade-offs between accuracy and responsiveness. We evaluated our system on 29 subjects via leave-one-subject-out cross-validation and obtained strong results, i.e., global accuracies of 70%, 85%, 89%, and 94% for the four classifiers operating at increasing timescales, respectively.

Game Theory for Multi-Access Edge Computing: Survey, Use Cases, and Future Trends

Game Theory (GT) has been used with significant success to formulate, and either design or optimize, the operation of many representative communications and networking scenarios. The games in these scenarios involve, as usual, diverse players with conflicting goals. This paper primarily surveys the literature that has applied theoretical games to wireless networks, emphasizing use cases of upcoming Multi-Access Edge Computing (MEC). MEC is relatively new and offers cloud services at the network periphery, aiming to reduce service latency backhaul load, and enhance relevant operational aspects such as Quality of Experience or security. Our presentation of GT is focused on the major challenges imposed by MEC services over the wireless resources. The survey is divided into classical and evolutionary games. Then, our discussion proceeds to more specific aspects which have a considerable impact on the game usefulness, namely: rational vs. evolving strategies, cooperation among players, available game information, the way the game is played (single turn, repeated), the game model evaluation, and how the model results can be applied for both optimizing resource-constrained resources and balancing diverse trade-offs in real edge networking scenarios. Finally, we reflect on lessons learned, highlighting future trends and research directions for applying theoretical model games in upcoming MEC services, considering both network design issues and usage scenarios.

Reliable Virtual Full-Duplex Relaying in the Presence of Interrelay Interference

Interrelay interference (IRI) is known to limit the performance of virtual full-duplex (VFD) relaying schemes. Available interference cancelation schemes are unable to simultaneously improve the diversity and maintain the multiplexing gain of VFD relaying schemes in multiple antennas scenarios. In this paper, we propose a technique to improve the reliability of multiantenna VFD relaying scheme, while maintaining the multiplexing gain in the presence of IRI. In the proposed scheme, the source and interrelay channels are concatenated and their signals are jointly decoded at the relay. A superposition of the decoded messages is forwarded to the destination. To reduce the effect of IRI on the receiving relay, two suboptimal relay transmit antenna selection schemes, namely, random antenna selection and maximum capacity antenna selection, are investigated. The system capacity and outage probability bounds of the proposed scheme are derived in closed form. The diversity and multiplexing tradeoff is also characterized. The analytical and numerical results show that the proposed scheme can achieve full diversity and maintain the multiplexing gain of the full-duplex relaying scheme. Also, the proposed scheme can outperform conventional VFD relaying scheme in terms of reliability even in the presence of IRI.

149 Human Subthalamic Nucleus Neurons Exhibit Increased Theta-band Phase-locking During High-conflict Decision Making

Abstract INTRODUCTION The subthalamic nucleus (STN) is thought to be preferentially engaged during high-conflict decision making in humans. The population neuronal spike rate in the STN has been reported to increase during decision conflict. Conflict and feedback-related activity is also reflected in theta-band (4-8 Hz) oscillations in the STN. It remains unknown how single-neuron activity and theta-band local field potentials (LFP) oscillations interact to support decision making. METHODS We simultaneously recorded single-neuron spike activity and LFP from the STN of eight Parkinson's disease (PD) patients while they performed a novel Aversion-Reward conflict (ARC) task. Subjects decide whether to accept an offer of a monetary reward paired with a variable risk of an aversive air puff to the eye. By varying the reward and risk, we are able to study approach-avoidance decision making across a range of conflict. Using this task, we examined the mechanism of how theta-frequency oscillation and entrained single neurons involve humans' integration of cost and benefit and decision at various conflict statuses. RESULTS >The ARC task reveals diverse risk-reward tradeoff strategies of patients. Consistent across patients, there is a positive correlation between the degree of decision conflict and reaction time (e.g., higher conflict offers require longer for subjects to decide). During high-conflict decisions, LFP in STN had increased activity of sub-theta oscillation, while increased activity of theta was found during low-conflict decisions. Single-trial STN theta-band power was correlated with degree of decision conflict. Interestingly, the decision to take or forgo the reward is predicted by theta-frequency phase-locked of STN neurons. CONCLUSION Our findings support the hypothesis that theta-band oscillations in single-neurons reflect the engagement of STN during conflict decision making. Furthermore, STN neurons with theta-band entrainment correlate with willingness to approach risk to pursue reward.

Analysis of opportunistic two‐path successive relaying in consideration of inter‐relay interference

Two-path successive relaying (TPSR) improves spectral efficiency of relay under half-duplex constraint. However, when the relay pair is operating in non-orthogonal channels, the capacity of TPSR can be severely degraded by inter-relay interference (IRI). Relay selection has been proposed in the existing literature to mitigate the IRI. However, existing opportunistic TPSR (OSR) schemes limit the pool of potential relay pair choices. Eventually, the performance of existing OSR schemes is underestimated. This study proposes four OSR schemes to improve the achievable performance of OSR schemes in terms of ergodic capacity and outage probability. The proposed schemes select a pair of relays jointly from the N potential relays. This increases the pool of potential relay pairs quadratically. Acknowledging the fact that OSR is only good for certain operating conditions, adaptive schemes that dynamically switches between OSR and opportunistic half-duplex relaying (OHR) are proposed. The simulation results show that the proposed schemes outperform the existing OSR, OHR and full-duplex relaying schemes significantly. The asymptotic ergodic capacity analysis of the proposed schemes increases logarithmically with the quadratic of N. The diversity and multiplexing tradeoff analysis reveals that the proposed schemes achieve the full diversity and multiplexing gain given a large number of potential relays.

Adaptive Cross-Generation Differential Evolution Operators for Multiobjective Optimization

Convergence performance and parametric sensitivity are two issues that tend to be neglected when extending differential evolution (DE) to multiobjective optimization (MO). To fill this research gap, we develop two novel mutation operators and a new parameter adaptation mechanism. A multiobjective DE variant is obtained through integration of the proposed strategies. The main innovation of this paper is the simultaneous use of individuals across generations from an objective-based perspective. Good convergence–diversity tradeoff and satisfactory exploration–exploitation balance are achieved via the hybrid cross-generation mutation operation. Furthermore, the cross-generation adaptation mechanism enables the individuals to self-adapt their associated parameters not only optimization-stage-wise but also objective-space-wise. Empirical results indicate the statistical superiority of the proposed algorithm over several state-of-the-art evolutionary algorithms in handling MO problems.

Performance Trade-Offs for Wavelength Striping Optical Switching Using a Novel Star Architecture

This work describes various performance trade-offs that arise from the use of a technique for optical switching under various network topologies. Such switching operation can be summarized as follows: (a) user data are divided into fixed-length fragments, (b) each fragment is assigned to a different wavelength, and (c) all wavelengths are simultaneously switched to the egress links. This concept of dividing user data into several wavelengths to be simultaneously switched is called wavelength striping and its purpose is to reduce latency and increase throughput for short distance interconnects. We depart from previous work where a building block implementing this basic switching function has been built around semiconductor optical amplifiers (SOAs). In this paper, we investigate diverse trade-offs that arise from the use of this switching approach in different network topologies. One of the main issues addressed in this paper is the relation between cascadability and bit error rate (BER). In this case, our results indicate that a switch fabric can cascade up to five stages without exceeding a BER of 10−9 and without incurring in power budget problems. We also show that the performance degradation, introduced by cascading SOAs, can be compensated with a star interconnect architecture that is introduced. Other issues addressed in this paper are the effect of scalability on cost and the effect of latency on TCP performance and reliability.

On Feasibility Conditions for Linear Interference Alignment over Cognitive Interference Channel: A DoF Analysis with Diversity Tradeoff

ABSTRACTIn a multiple-antennas cognitive radio network, the rate of the primary link is maximized by water-filling power allocation along the spatial directions associated with the singular values of primary channel matrix. Due to transmitter power constraint, even when the primary user occupies the entire spectrum, some spatial opportunities remain that can be utilized by secondary users. This scheme is called opportunistic interference alignment (OIA). The feasibility issue of OIA in a general symbol-extended system considering the diversity tradeoff, in particular, is the topic that is considered in this paper. An outer bound on the achievable degrees of freedom and also a necessary condition on the feasibility of a multiuser OIA network are presented analytically. In the remaining parts of the paper, the diversity and rate tradeoff is considered. In order to achieve diversity, it is assumed that the paired nodes utilize a space-time coder before the OIA pre-coder. A closed-form relation is presented that relates the minimum achievable diversity to interference alignment quality and code rate.

Evaluating taboo trade-offs in ecosystems services and human well-being

Managing ecosystems for multiple ecosystem services and balancing the well-being of diverse stakeholders involves different kinds of trade-offs. Often trade-offs involve noneconomic and difficult-to-evaluate values, such as cultural identity, employment, the well-being of poor people, or particular species or ecosystem structures. Although trade-offs need to be considered for successful environmental management, they are often overlooked in favor of win-wins. Management and policy decisions demand approaches that can explicitly acknowledge and evaluate diverse trade-offs. We identified a diversity of apparent trade-offs in a small-scale tropical fishery when ecological simulations were integrated with participatory assessments of social-ecological system structure and stakeholders' well-being. Despite an apparent win-win between conservation and profitability at the aggregate scale, food production, employment, and well-being of marginalized stakeholders were differentially influenced by management decisions leading to trade-offs. Some of these trade-offs were suggested to be "taboo" trade-offs between morally incommensurable values, such as between profits and the well-being of marginalized women. These were not previously recognized as management issues. Stakeholders explored and deliberated over trade-offs supported by an interactive "toy model" representing key system trade-offs, alongside qualitative narrative scenarios of the future. The concept of taboo trade-offs suggests that psychological bias and social sensitivity may exclude key issues from decision making, which can result in policies that are difficult to implement. Our participatory modeling and scenarios approach has the potential to increase awareness of such trade-offs, promote discussion of what is acceptable, and potentially identify and reduce obstacles to management compliance.

The Emerging Soybean Production Frontier in Southern Africa: Conservation Challenges and the Role of South-South Telecouplings

Soybean expansion has been a strong driver of deforestation and biodiversity loss in South America. Here we highlight strong similarities in environmental, institutional, and other contextual conditions among South American (SAM) and Southern African (SAFR) dry forest and savanna regions, and compile evidence for an emerging soybean production frontier in Southern Africa. Knowledge transfer, cooperation, and direct investment between South American and Southern African countries constitute crucial elements of soybean expansion in Africa. Comparing maps of soybean suitability, biodiversity, and carbon revealed substantial and spatially diverse trade-offs, suggesting that the emerging soybean frontier in Southern Africa may poses major challenges for conservation. An increased focus of conservation science on agricultural expansion and intensification in Southern African, as well as strong environmental policies for balancing agricultural production and conservation goals, are needed to mitigate potentially large trade-offs. The coupling of production frontiers should be a vehicle for the transfer not only of agricultural technology and production models, but also of experiences in environmental governance on emerging agricultural frontiers.

An effective multiobjective approach for hard partitional clustering

Clustering is an unsupervised classification method in the field of data mining. Many population based evolutionary and swarm intelligence optimization methods are proposed to optimize clustering solutions globally based on a single selected objective function which lead to produce a single best solution. In this sense, optimized solution is biased towards a single objective, hence it is not equally well to the data set having clusters of different geometrical properties. Thus, clustering having multiple objectives should be naturally optimized through multiobjective optimization methods for capturing different properties of the data set. To achieve this clustering goal, many multiobjective population based optimization methods, e.g., multiobjective genetic algorithm, mutiobjective particle swarm optimization (MOPSO), are proposed to obtain diverse tradeoff solutions in the pareto-front. As single directional diversity mechanism in particle swarm optimization converges prematurely to local optima, this paper presents a two-stage diversity mechanism in MOPSO to improve its exploratory capabilities by incorporating crossover operator of the genetic algorithm. External archive is used to store non-dominated solutions, which is further utilized to find one best solution having highest F-measure value at the end of the run. Two conceptually orthogonal internal measures SSE and connectedness are used to estimate the clustering quality. Results demonstrate effectiveness of the proposed method over its competitors MOPSO, non-dominated sorting genetic algorithm, and multiobjective artificial bee colony on seven real data sets from UCI machine learning repository.

Pruning of Error Correcting Output Codes by optimization of accuracy–diversity trade off

Ensemble learning is a method of combining learners to obtain more reliable and accurate predictions in supervised and unsupervised learning. However, the ensemble sizes are sometimes unnecessarily large which leads to additional memory usage, computational overhead and decreased effectiveness. To overcome such side effects, pruning algorithms have been developed; since this is a combinatorial problem, finding the exact subset of ensembles is computationally infeasible. Different types of heuristic algorithms have developed to obtain an approximate solution but they lack a theoretical guarantee. Error Correcting Output Code (ECOC) is one of the well-known ensemble techniques for multiclass classification which combines the outputs of binary base learners to predict the classes for multiclass data. In this paper, we propose a novel approach for pruning the ECOC matrix by utilizing accuracy and diversity information simultaneously. All existing pruning methods need the size of the ensemble as a parameter, so the performance of the pruning methods depends on the size of the ensemble. Our unparametrized pruning method is novel as being independent of the size of ensemble. Experimental results show that our pruning method is mostly better than other existing approaches.