Fairness Constraints Research Articles

Kernel dependence regularizers and Gaussian processes with applications to algorithmic fairness

Current adoption of machine learning in industrial, societal and economical activities has raised concerns about the fairness, equity and ethics of automated decisions. Predictive models are often developed using biased datasets and thus retain or even exacerbate biases in their decisions and recommendations. Removing the sensitive covariates, such as gender or race, is insufficient to remedy this issue since the biases may be retained due to other related covariates. We present a regularization approach to this problem that trades off predictive accuracy of the learned models (with respect to biased labels) for the fairness in terms of statistical parity, i.e. independence of the decisions from the sensitive covariates. In particular, we consider a general framework of regularized empirical risk minimization over reproducing kernel Hilbert spaces and impose an additional regularizer of dependence between predictors and sensitive covariates using kernel-based measures of dependence, namely the Hilbert-Schmidt Independence Criterion (HSIC) and its normalized version. This approach leads to a closed-form solution in the case of squared loss, i.e. ridge regression. We also provide statistical consistency results for both risk and fairness bound for our approach. Moreover, we show that the dependence regularizer has an interpretation as modifying the corresponding Gaussian process (GP) prior. As a consequence, a GP model with a prior that encourages fairness to sensitive variables can be derived, allowing principled hyperparameter selection and studying of the relative relevance of covariates under fairness constraints. Experimental results in synthetic examples and in real problems of income and crime prediction illustrate the potential of the approach to improve fairness of automated decisions.

An Annual Electric Energy Trade Scheduling Model under the Dual Track Mode

The annual electricity trade scheduling is the basis of long-term power generation scheduling. In recent decades, the ratio of new energy generation in China has increased annually, and the electricity market has operated under the “market electricity” and “planned electricity” double track mode in recent years. However, the existing annual electricity trade scheduling methods are extensive and cannot adapt to the new situation of “market electricity” and large-scale new energy generation. The annual scheduled energy of the power units is set as a decision variable, and a novel annual energy scheduling optimization model based on Gini coefficient of fairness is presented in this paper. In this model, “market electricity” capacity is conversed monthly, considering peaking reserve demand and monthly characteristics of new energy generation. The fairness constraint set based on Gini coefficient is introduced into the optimization model to solve various fairness problems. Simulation results show that the introduction of the Gini coefficient and the optimization model considering the monthly conversion of marketing electricity capacity can obtain more accurate and reasonable electricity distribution results, and the peaking demand can be considered more fairly and effectively. The proposed method provides a feasible solution to the annual electric energy scheduling for dual-track operation country such as China.

A Local Search Framework for Experimental Design

We present a local search framework to design and analyze both combinatorial algorithms and rounding algorithms for experimental design problems. This framework provides a unifying approach to match and improve all known results in D/A/E-design and to obtain new results in previously unknown settings. For combinatorial algorithms, we provide a new analysis of the classical Fedorov's exchange method. We prove that this simple local search algorithm works well as long as there exists an almost optimal solution with good condition number. Moreover, we design a new combinatorial local search algorithm for E-design using the regret minimization framework. For rounding algorithms, we provide a unified randomized exchange algorithm to match and improve previous results for D/A/E-design. Furthermore, the algorithm works in the more general setting to approximately satisfy multiple knapsack constraints, which can be used for weighted experimental design and for incorporating fairness constraints into experimental design.

Maximizing fair content spread via edge suggestion in social networks

Content spread inequity is a potential unfairness issue in online social networks, disparately impacting minority groups. In this paper, we view friendship suggestion, a common feature in social network platforms, as an opportunity to achieve an equitable spread of content. In particular, we propose to suggest a subset of potential edges (currently not existing in the network but likely to be accepted) that maximizes content spread while achieving fairness. Instead of re-engineering the existing systems, our proposal builds a fairness wrapper on top of the existing friendship suggestion components. We prove the problem is NP-hard and inapproximable in polynomial time unless P = NP. Therefore, allowing relaxation of the fairness constraint, we propose an algorithm based on LP-relaxation and randomized rounding with fixed approximation ratios on fairness and content spread. We provide multiple optimizations, further improving the performance of our algorithm in practice. Besides, we propose a scalable algorithm that dynamically adds subsets of nodes, chosen via iterative sampling, and solves smaller problems corresponding to these nodes. Besides theoretical analysis, we conduct comprehensive experiments on real and synthetic data sets. Across different settings, our algorithms found solutions with near-zero unfairness while significantly increasing the content spread. Our scalable algorithm could process a graph with half a million nodes on a single machine, reducing the unfairness to around 0.0004 while lifting content spread by 43%.

A Framework for an Efficient Recommendation System Using Time and Fairness Constraint Based Web Usage Mining Technique

Users prefer to use various websites like Facebook, Gmail, and YouTube. We can make the system predict what pages we expect in the future and give the users what they have requested. Based on the data gathered and analyzed, we can predict the user's future navigation patterns in response to the user's requests. In order to track down users’ navigational sessions, the web access logs created at a specific website are processed. Grouping the user session data is then done into clusters, where inter-cluster similarities are minimized, although the intra-cluster similarities are maximised. Recent clustering and fairness analysis research has focused on centric-based methods such as k-median and k-means clustering. We propose improved constrained based clustering (ICBC) based on fair algorithms for managing Hierarchical Agglomerative Clustering (HAC) that apply fairness constraints regardless of distance linking parameters, simplifying clustering fairness trials for HAC and intended for various protected groups compared to vanilla HAC techniques. Also, this ICBC is used to select an algorithm whose inherent bias matches a specific problem, and then to adjust the optimization criterion of any distinct algorithm to take the constraints on interpretation to improve the efficiency of clustering. We show that our proposed algorithm finds fairer clustering by evaluation on the NASA dataset by balancing the constraints of the problem.

Identity and the limits of fair assessment

In many assessment problems—aptitude testing, hiring decisions, appraisals of the risk of recidivism, evaluation of the credibility of testimonial sources, and so on—the fair treatment of different groups of individuals is an important goal. But individuals can be legitimately grouped in many different ways. Using a framework and fairness constraints explored in research on algorithmic fairness, I show that eliminating certain forms of bias across groups for one way of classifying individuals can make it impossible to eliminate such bias across groups for another way of dividing people up. And this point generalizes if we require merely that assessments be approximately bias-free. Moreover, even if the fairness constraints are satisfied for some given partitions of the population, the constraints can fail for the coarsest common refinement, that is, the partition generated by taking intersections of the elements of these coarser partitions. This shows that these prominent fairness constraints admit the possibility of forms of intersectional bias.

On the Impossibility of Non-trivial Accuracy in Presence of Fairness Constraints

One of the main concerns about fairness in machine learning (ML) is that, in order to achieve it, one may have to trade off some accuracy. To overcome this issue, Hardt et al. proposed the notion of equality of opportunity (EO), which is compatible with maximal accuracy when the target label is deterministic with respect to the input features. In the probabilistic case, however, the issue is more complicated: It has been shown that under differential privacy constraints, there are data sources for which EO can only be achieved at the total detriment of accuracy, in the sense that a classifier that satisfies EO cannot be more accurate than a trivial (random guessing) classifier. In our paper we strengthen this result by removing the privacy constraint. Namely, we show that for certain data sources, the most accurate classifier that satisfies EO is a trivial classifier. Furthermore, we study the trade-off between accuracy and EO loss (opportunity difference), and provide a sufficient condition on the data source under which EO and non-trivial accuracy are compatible.

Group-Aware Threshold Adaptation for Fair Classification

The fairness in machine learning is getting increasing attention, as its applications in different fields continue to expand and diversify. To mitigate the discriminated model behaviors between different demographic groups, we introduce a novel post-processing method to optimize over multiple fairness constraints through group-aware threshold adaptation. We propose to learn adaptive classification thresholds for each demographic group by optimizing the confusion matrix estimated from the probability distribution of a classification model output. As we only need an estimated probability distribution of model output instead of the classification model structure, our post-processing model can be applied to a wide range of classification models and improve fairness in a model-agnostic manner and ensure privacy. This even allows us to post-process existing fairness methods to further improve the trade-off between accuracy and fairness. Moreover, our model has low computational cost. We provide rigorous theoretical analysis on the convergence of our optimization algorithm and the trade-off between accuracy and fairness. Our method theoretically enables a better upper bound in near optimality than previous method under the same condition. Experimental results demonstrate that our method outperforms state-of-the-art methods and obtains the result that is closest to the theoretical accuracy-fairness trade-off boundary.

Modification-Fair Cluster Editing

The classic Cluster Editing problem (also known as Correlation Clustering) asks to transform a given graph into a disjoint union of cliques (clusters) by a small number of edge modifications. When applied to vertex-colored graphs (the colors representing subgroups), standard algorithms for the NP-hard Cluster Editing problem may yield solutions that are biased towards subgroups of data (e.g., demographic groups), measured in the number of modifications incident to the members of the subgroups. We propose a modification fairness constraint which ensures that the number of edits incident to each subgroup is proportional to its size. To start with, we study Modification-Fair Cluster Editing for graphs with two vertex colors. We show that the problem is NP-hard even if one may only insert edges within a subgroup; note that in the classic "non-fair" setting, this case is trivially polynomial-time solvable. However, in the more general editing form, the modification-fair variant remains fixed-parameter tractable with respect to the number of edge edits. We complement these and further theoretical results with an empirical analysis of our model on real-world social networks where we find that the price of modification-fairness is surprisingly low, that is, the cost of optimal modification-fair differs from the cost of optimal "non-fair" solutions only by a small percentage.

Flexible FOND Planning with Explicit Fairness Assumptions


 
 
 We consider the problem of reaching a propositional goal condition in fully-observable nondeterministic (FOND) planning under a general class of fairness assumptions that are given explicitly. The fairness assumptions are of the form A/B and say that state trajectories that contain infinite occurrences of an action a from A in a state s and finite occurrence of actions from B, must also contain infinite occurrences of action a in s followed by each one of its possible outcomes. The infinite trajectories that violate this condition are deemed as unfair, and the solutions are policies for which all the fair trajectories reach a goal state. We show that strong and strong-cyclic FOND planning, as well as QNP planning, a planning model introduced recently for generalized planning, are all special cases of FOND planning with fairness assumptions of this form which can also be combined. FOND+ planning, as this form of planning is called, combines the syntax of FOND planning with some of the versatility of LTL for expressing fairness constraints. A sound and complete FOND+ planner is implemented by reducing FOND+ planning to answer set programs, and its performance is evaluated in comparison with FOND and QNP planners, and LTL synthesis tools. Two other FOND+ planners are introduced as well which are more scalable but are not complete.
 
 
 

Path Planning for the Dynamic UAV-Aided Wireless Systems Using Monte Carlo Tree Search

For UAV-aided wireless systems, online path planning attracts much attention recently. To better adapt to the real-time dynamic environment, for the first time, we propose a Monte Carlo Tree Search (MCTS)-based path planning scheme. In details, we consider a single UAV acts as a mobile server to provide computation tasks offloading services for a set of mobile users on the ground, where the movement of ground users follows a Random Way Point model. Our model aims at maximizing the average throughput under energy consumption and user fairness constraints, and the proposed time-saving MCTS algorithm can further improve the performance. Simulation results show that the proposed algorithm achieves a larger average throughput and a faster convergence performance compared with the baseline algorithms of Q-learning and Deep Q-Network.

Approximate and Deployable Shortest Remaining Processing Time Scheduler

The scheduling policy installed on switches of datacenters plays a significant role on congestion control. Shortest-Remaining-Processing-Time (SRPT) achieves the near-optimal average message completion time (MCT) in various scenarios, but is difficult to deploy as viewed by the industry. The reasons are two-fold: 1) many commodity switches only provide FIFO queues, and 2) the information of remaining message size is not available. Recently, the idea of emulating SRPT using only a few FIFO queues and the original message size has been coined as the approximate and deployable SRPT (ADS) design. In this paper, we provide the first theoretical study on the optimal ADS design. Specifically, we first characterize a wide range of feasible ADS scheduling policies via a unified framework, and then derive the steady-state MCT, slowdown, and impoliteness in the M/G/1 setting. Hence we formulate the optimal ADS design as a non-linear combinatorial optimization problem, which aims to minimize the average MCT given the available FIFO queues. We also take into account the proportional fairness and temporal fairness constraints based on the maximal slowdown and impoliteness, respectively. The optimal ADS design problem is NP-hard in general, and does not exhibit monotonicity or sub-modularity. We leverage its decomposable structure and devise an efficient algorithm to solve the optimal ADS policy. We carry out extensive flow-level simulations and packet-level experiments to evaluate the proposed optimal ADS design. Results show that the optimal ADS policy installed on eight FIFO queues is capable of emulating the true SRPT.

TabFairGAN: Fair Tabular Data Generation with Generative Adversarial Networks

With the increasing reliance on automated decision making, the issue of algorithmic fairness has gained increasing importance. In this paper, we propose a Generative Adversarial Network for tabular data generation. The model includes two phases of training. In the first phase, the model is trained to accurately generate synthetic data similar to the reference dataset. In the second phase we modify the value function to add fairness constraint, and continue training the network to generate data that is both accurate and fair. We test our results in both cases of unconstrained, and constrained fair data generation. We show that using a fairly simple architecture and applying quantile transformation of numerical attributes the model achieves promising performance. In the unconstrained case, i.e., when the model is only trained in the first phase and is only meant to generate accurate data following the same joint probability distribution of the real data, the results show that the model beats the state-of-the-art GANs proposed in the literature to produce synthetic tabular data. Furthermore, in the constrained case in which the first phase of training is followed by the second phase, we train the network and test it on four datasets studied in the fairness literature and compare our results with another state-of-the-art pre-processing method, and present the promising results that it achieves. Comparing to other studies utilizing GANs for fair data generation, our model is comparably more stable by using only one critic, and also by avoiding major problems of original GAN model, such as mode-dropping and non-convergence.

다중 파라메트릭 변환곡선 기반 선수 선형 변환기법 연구

In this paper, we propose a systematic hull form variation technique which automatically satisfies the displacement constraint and guarantees a high level of fairness. This method is possible through multiple parameter correction curves. The present method is to improve the hull form variation method based on parametric modification function and consists of two sub-categories: SAC variation and section lines modification. For SAC variation, the utilization of two B-Spline curves satisfying GC¹ condition led to the satisfaction of displacement constraint and high level of fairness at the same time. Section lines modification methods involves in using two fuctions: the first is the waterplane modification function combining two cubic splines. the other function is the sectional area modification function consisting of 2nd order polynomial over the DLWL(Design Load Waterline) and 3rd order polynomial below the DLWL, This function enables not only the fundamental U-V section shape variation but also systematically modified section lines. The present method is expected to be more useful in the hull form optimization process using CFD compared to the existing method.

Low Complexity Joint Sub-Carrier Pairing, Allocation and Relay Selection in Cooperative Wireless Networks

Multi-carrier cooperative relay-based wireless communication is of particular interest in future wireless networks. In this paper we present resource allocation algorithm in which sub-carrier pairing is of particular interest along with fairness constraint in multi-user networks. An optimization of sub-carrier pair selection is formulated through capacity maximization problem. Sub-carrier pairing is applied in both two-hop Amplify & Forward (AF) and Decode & Forward (DF) cooperative multi-user networks. We develop a less complex centralized scheme for joint Sub-carrier pairing and allocation along with relay selection. The computational complexity of the proposed algorithms has been analyzed and performance is compared with Exhaustive Search Algorithm.

QoS-Driven Load Balancing in Hybrid LiFi and WiFi Networks

This work studies the quality of service (QoS) performance of wireless networks that integrate light fidelity (LiFi) and wireless fidelity (WiFi). While the hybrid network is potential for improving network capacity, load balancing becomes essential and challenging due to the nature of heterogeneous access points (APs). A number of studies have been conducted to address this issue, focusing on maximising the network capacity with user fairness constraints. However, in practice, QoS metrics including packet loss ratio and latency are important to network services. In this paper, QoS-driven load balancing is studied for hybrid LiFi and WiFi networks (HLWNets) in two scenarios: single-AP association (SA) and multi-AP association (MA). In each case, an optimisation problem is formulated to minimise the packet loss ratio and latency, and a low-complexity iterative algorithm is proposed to solve the problem. Results show that the novel methods, especially MA, can effectively balance the traffic loads among the APs and improve the QoS performance. In addition, the more subflows the better performance MA provides. Targeting the same level of QoS, MA can achieve a system throughput up to 160% higher than the signal strength strategy and 130% higher than the proportional fairness load balancing.

Evaluating algorithmic fairness in the presence of clinical guidelines: the case of atherosclerotic cardiovascular disease risk estimation

ObjectivesThe American College of Cardiology and the American Heart Association guidelines on primary prevention of atherosclerotic cardiovascular disease (ASCVD) recommend using 10-year ASCVD risk estimation models to initiate statin treatment....

A Maximal Correlation Framework for Fair Machine Learning.

As machine learning algorithms grow in popularity and diversify to many industries, ethical and legal concerns regarding their fairness have become increasingly relevant. We explore the problem of algorithmic fairness, taking an information–theoretic view. The maximal correlation framework is introduced for expressing fairness constraints and is shown to be capable of being used to derive regularizers that enforce independence and separation-based fairness criteria, which admit optimization algorithms for both discrete and continuous variables that are more computationally efficient than existing algorithms. We show that these algorithms provide smooth performance–fairness tradeoff curves and perform competitively with state-of-the-art methods on both discrete datasets (COMPAS, Adult) and continuous datasets (Communities and Crimes).

Bidding behaviors of GENCOs under bounded rationality with renewable energy

With the worldwide restructuring of power markets, it is more meaningful to analyze the bidding behaviors of generation companies (GENCOs). A realistic bidding behavior model can help to design better power market mechanisms. However, most existing studies usually use strong assumptions, where all GENCOs behave perfectly rationally when participating in power markets. Many GENCOs' observed bidding behaviors are proved to deviate from the simulated perfectly rational ones. Thus, it is significant to introduce individual subjective perspectives into the bidding behavior modeling to obtain more realistic models. Although several studies have been working on this topic by introducing bounded rationality, they still have some problems in modeling bounded rationality, power system operation, power market rules, etc. This paper proposes a novel bidding behavior model for GENCOs under bounded rationality with renewable energy to overcome these problems. The introduced bounded rationality features include prospect theory in the form of framing effects and the concept of fairness constraints on profit-making. A Stackelberg game model is formulated to analyze the interaction between a GENCO with renewables and a day-ahead power market. The proposed model is nonlinear, which is further transformed into a mixed-integer linear formulation by applying Karush–Kuhn–Tucker theorem and strong duality theorem. Based on a modified IEEE-30 bus system with high renewable, an illustrative example is proposed to demonstrate the effectiveness of the proposed model and analyze the bidding behaviors of bounded rational GENCOs and the corresponding effects on power markets.

Price Discrimination with Fairness Constraints

Price discrimination strategies, which offer different prices to customers based on differences in their valuations, have become common practice. Although it allows sellers to increase their profits, it also raises several concerns in terms of fairness (e.g., by charging higher prices (or denying access) to protected minorities in case they have higher (or lower) valuations than the general population). This topic has received extensive attention from media, industry, and regulatory agencies. In this paper, we consider the problem of setting prices for different groups under fairness constraints. We first propose four definitions: fairness in price, demand, consumer surplus, and no-purchase valuation. We prove that satisfying more than one of these fairness constraints is impossible even under simple settings. We then analyze the pricing strategy of a profit-maximizing seller and the impact of imposing fairness on the seller’s profit, consumer surplus, and social welfare. Under a linear demand model, we find that imposing a small amount of price fairness increases social welfare, whereas too much price fairness may result in a lower welfare relative to imposing no fairness. On the other hand, imposing fairness in demand or consumer surplus always decreases social welfare. Finally, no-purchase valuation fairness always increases social welfare. We observe similar patterns under several extensions and for other common demand models numerically. Our results and insights provide a first step in understanding the impact of imposing fairness in the context of discriminatory pricing. This paper was accepted by Jayashankar Swaminathan, operations management. Funding: A. N. Elmachtoub was supported by the Division of Civil, Mechanical and Manufacturing Innovation [Grants 1763000 and 1944428]. Supplemental Material: The data files and online appendix are available at https://doi.org/10.1287/mnsc.2022.4317 .