Machine Learning (ML) algorithms are increasingly used in decision-making process across various social-critical domains, but they often somewhat inherit and amplify bias from their training data, leading to unfair and unethical outcomes. This issue highlights the urgent need for effective methods to detect, explain, and mitigate bias to ensure the fairness of ML systems. Previous studies are prone to analyze the root causes of algorithmic bias from a statistical perspective. However, to the best of our knowledge, none of them has discussed how sensitive information inducing the final discriminatory decision is encoded by ML models. In this work, we attempt to explain and mitigate algorithmic bias from a game-theoretic view. We mathematically decode an essential and common component of sensitive information implicitly defined by various fairness metrics with Harsanyi interactions, and on this basis, we propose an in-processing method HIFI for bias mitigation. We conduct an extensive evaluation of HIFI with 11 state-of-the-art methods, 5 real-world datasets, 4 fairness criteria, and 5 ML performance metrics, while also considering intersectional fairness for multiple protected attributes. The results show that HIFI surpasses state-of-the-art in-processing methods in terms of fairness improvement and fairness-performance trade-off, and also achieves notable effectiveness in reducing violations of individual fairness simultaneously.

Deep reinforcement learning (DRL) has emerged as a promising approach to address challenges in autonomous driving. Recently, reinforcement learning from expert demonstrations has gained considerable attention, offering a synergistic integration of imitation learning (IL) and reinforcement learning to enhance model performance. However, in complex driving scenarios, expert decisions may be suboptimal, leading to inefficient training and susceptibility to local optima. In light of this, we propose a novel DRL-based model, termed PME. In the decision-making process, we introduce an enhanced algorithm built upon Proximal Policy Optimization (PPO), which integrates multiple expert policies to mitigate the negative impact of erroneous expert decisions, thereby significantly improving the overall performance of the autonomous driving model. Experimental results, conducted in the CARLA simulator, demonstrate that the proposed PME model surpasses conventional reinforcement learning algorithms across various driving scenarios. The framework consistently achieves higher success rates under diverse conditions, showcasing improved learning stability and superior generalization capabilities.

Machine learning learns patterns from data to improve the performance of the decision-making systems through computing, and gradually affects people’s lives. However, it shows that in current research machine learning algorithms may reinforce human discrimination, and exacerbate negative impacts on unprivileged groups. To mitigate potential unfairness in machine learning classifiers, we propose a fair classification approach by quantifying the difference in the prediction distribution with the idea of correlation alignment in transfer learning, which improves fairness efficiently by minimizing the second-order statistical distance of the prediction distribution. We evaluate the validity of our approach on four real-world datasets. It demonstrates that our approach significantly mitigates bias w.r.t demographic parity, equality of opportunity, and equalized odds across different groups in a classification setting, and achieves better trade-off between accuracy and fairness than previous work. In addition, our approach can further improve fairness and mitigate the fair conflict problem in debiased networks.

Deep neural networks (DNNs) have shown powerful performance in various applications and are increasingly being used in decisionmaking systems. However, concerns about fairness in DNNs always persist. Some efficient white-box fairness testing methods about individual fairness have been proposed. Nevertheless, the development of black-box methods has stagnated, and the performance of existing methods is far behind that of white-box methods. In this paper, we propose a novel black-box individual fairness testing method called Model-Agnostic Fairness Testing (MAFT). By leveraging MAFT, practitioners can effectively identify and address discrimination in DL models, regardless of the specific algorithm or architecture employed. Our approach adopts lightweight procedures such as gradient estimation and attribute perturbation rather than non-trivial procedures like symbol execution, rendering it significantly more scalable and applicable than existing methods. We demonstrate that MAFT achieves the same effectiveness as state-of-the-art white-box methods whilst improving the applicability to large-scale networks. Compared to existing black-box approaches, our approach demonstrates distinguished performance in discovering fairness violations w.r.t effectiveness (~ 14.69×) and efficiency (~ 32.58×).

As a rapidly evolving AI technology, deep neural networks are becoming increasingly integrated into human society, yet raising concerns about fairness issues. Previous studies have proposed a metric called causal fairness to measure the fairness of machine learning models and proposed some search algorithms to mine individual discrimination instance pairs (IDIPs). Fairness issues can be alleviated by retraining models with corrected IDIPs. However, the number of samples that are used as seeds for these methods is often limited due to the pursuit of efficiency. In addition, the quantity of IDIPs generated on different seeds varies, so it makes sense to select appropriate samples as seeds, which has not been sufficiently considered in past studies. In this paper, we study the imbalance in IDIP quantities for various datasets and sensitive attributes, highlighting the need for selecting and ranking seed samples. Then, we proposed FIPSER, a feature importance and perturbation potential-based seed prioritization method. Our experimental results show that, on average, when applied to the current state-of-the-art method of IDIP mining, FIPSER can improve its effectiveness by 45% and efficiency by 11%.

Deep learning (DL) systems are increasingly deployed for autonomous decision-making in a wide range of applications. Apart from the robustness and safety, fairness is also an important property that a well-designed DL system should have. To evaluate and improve individual fairness of a model, systematic test case generation for identifying individual discriminatory instances in the input space is essential. In this paper, we propose a framework EIDIG for efficiently discovering individual fairness violation. Our technique combines a global generation phase for rapidly generating a set of diverse discriminatory seeds with a local generation phase for generating as many individual discriminatory instances as possible around these seeds under the guidance of the gradient of the model output. In each phase, prior information at successive iterations is fully exploited to accelerate convergence of iterative optimization or reduce frequency of gradient calculation. Our experimental results show that, on average, our approach EIDIG generates 19.11% more individual discriminatory instances with a speedup of 121.49% when compared with the state-of-the-art method and mitigates individual discrimination by 80.03% with a limited accuracy loss after retraining.

Open Automata (OA) are symbolic and parameterized models for open concurrent systems, where open means partially specified systems, which can be instantiated or assembled to build bigger systems. In previous work, a notion of equivalence named FH-Bisimulation was defined for OA, coming with both Strong and Weak flavors, where Weak means ignoring internal moves when they do not affect the external behavior. Both flavors have been proven to be congruent for the OA’s composition. In this paper, we propose a new definition of (weak) OA, that is both more expressive for encoding the behavior of parameterized systems and suitable as a finite encoding of weak OA. We name this meta (weak) OA and provide two methods to check their equivalence, either explicitly building the meta-WOA, or constructing their meta open transitions on-demand. The last strategy has better termination properties.

Backbone is the common part of each solution in a given propositional formula, which is a key to improving the performance of SAT solving and SAT-based applications, such as model checking and program analysis. In this paper, we propose an optimized approach that combines implication-driven (IDF), conflict-driven (CDF), and unique-driven (UDF) heuristics to improve backbone computing. IDF uses the particular binary structure of the form a ↔ b ∧ c to find more backbone literals. CDF comes from the observation that for a clause ¬a V b, if a is a backbone literal, then b is also a backbone literal. Besides CDF, we are also able to detect new non-backbone literals by UDF. A literal l is not a backbone literal, if there is no clause Φ ϵ Φ that is only satisfied by l. We implemented our approach in a tool named DUCIBone with the above optimizations (IDF+CDF+UDF), and conducted experiments on formulas used in previous work and SAT competitions (2015, 2016). Results demonstrate that DUCIBone solved 4% (507 formulas) more formulas than minibones (minibones-RLD, 490 formulas) does under its best configuration. Among 486 formulas solved by all tools (DUCIBone, minibones-RLD, minibonescb100), DUCIBone reduced 7% (35131 seconds) than minibones (37454 seconds). Experiments indicate that the advantage of DUCIBone is more obvious when the formulas are harder.

Summary Recommender systems are playing an increasing role in improving user satisfaction as they can recommend items which might be highly interested to users. Recent advances have proven that social relations such as trust and distrust relations among users are helpful in improving recommendation accuracy. Traditional social recommendation methods directly utilize unweighted trust and distrust relations into collaborative filtering framework. These methods will lose their power when the trust or distrust relation data is sparse, which significantly hinders the improvement of rating prediction accuracy. To address this problem, we transform the unweighted trust and distrust relations into fine‐grained weighted social trust matrix which is denser and encodes the trust and distrust degree for pair of users. The weighted social trust matrix is then combined with the rating matrix in a collective matrix factorization framework to implement rating prediction task. Experimental results based on Extended Epinions dataset show that the proposed collective matrix factorization model with fine‐grained weighted social trust matrix can achieve better accuracy than conventional social recommendation algorithms such as SoRec and its extensions.

Probabilistic Model-Checking is a standard approach for automatically verifying stochastic systems. However, it becomes expensive or even intractable for classic approaches to verify complex systems. Statistical model-checking was proposed to overcome this limitation. In this paper, we propose a novel statistical model-checking approach which is based on Bayesian point estimation. Together with the Bayesian point estimation and a given conjugate prior distribution, we are able to predict the upper bound of sample size before sampling. We implement our techniques in a tool. Experiential results show that our approach is competitive, even better than other standard approaches in several cases.

Pushdown systems with transductions (TrPDSs) are an extension of pushdown systems (PDSs) by associating each transition rule with a transduction, which allows to inspect and modify the stack content at each step of a transition rule. It was shown by Uezato and Minamide that TrPDSs can model PDSs with checkpoint and discrete-timed PDSs. Moreover, TrPDSs can be simulated by PDSs and the predecessor configurations pre^*(C) of a regular set C of configurations can be computed by a saturation procedure when the closure of the transductions in TrPDSs is finite. In this work, we comprehensively investigate the reachability problem of finite TrPDSs. We propose a novel saturation procedure to compute pre^*(C) for finite TrPDSs. Also, we introduce a saturation procedure to compute the successor configurations post^*(C) of a regular set C of configurations for finite TrPDSs. From these two saturation procedures, we present two efficient implementation algorithms to compute pre^*(C) and post^*(C). Finally, we show how the presence of transductions enables the modeling of Boolean programs with call-by-reference parameter passing. The TrPDS model has finite closure of transductions which results in model-checking approach for Boolean programs with call-by-reference parameter passing against safety properties.

Pipe protocol, proposed by Zhao [1] in early 2013, is one application layer protocol and one way to establish the Internet of Things, under which can different kinds of hardware platforms communicate with each other faster and more safely. As an upper layer protocol, Pipe protocol doesn't define details, but during the implementation, probabilistic and nondeterministic behaviors, such as data loss and external choice, are possible to happen. In this paper, we use probabilistic model checker, PRISM, to construct the probabilistic Pipe protocol as Probabilistic Timed Automata (PTAs), then verify some useful time-bounded properties, written in Probabilistic Computation Tree Logic (PCTL), like, “Maximum probability that the pipe is shut down after Source node sends all 5 data packets within 50 μs”. The results show that the data loss probabilities, number of data and deadline should be restricted suitably if we require the maximum probability of the goal reaches some value. Our model is proved to be meaningful and we can give helpful suggestions to improve the implementation of Pipe protocol.

To model and analyze systems with multi-valued information, in this paper, we present an extension of Kripke structures in the framework of complete residuted lattices, which we will refer to as lattice-valued Kripke structures (LKSs). We then show how the traditional trace containment and equivalence relations, can be lifted to the lattice-valued setting, and we introduce two families of lattice-valued versions of the relations. Further, we explore some interesting properties of these relations. Finally, we provide logical characterizations of our relations by a natural extension of linear temporal logic.

In order to describe the requirement of spatial and temporal consistency of cyber-physical systems, a specification language called as STeC was proposed by Chen in [1]. In this paper, we focus on the theory of semantics of STeC. After simply restating the syntax and operational semantics, we mainly establish the denotational semantics of STeC. To investigate the reasonability of the denotational semantics, an abstract theorem is given to show the soundness and completeness of the denotational semantics. Finally, a simple case about China Gaotie (which means High-speed train) is given to show how to compute the operational and denotational semantics.

In this paper, we define lattice-valued labeled transition systems (LLTS) as a general framework for allowing imprecise or incomplete specifications to be expressed. We introduce a lattice-valued bisimulation between LLTSs that measures the degree of closeness of two systems as elements of residuated lattice, in contrast to the traditional boolean yes/no to bisimulation. Also, we show that our bisimulation is compositional for a synchronous composition operator. Moreover, we also consider lattice-valued extension of Kripke structures, define a lattice-valued bisimulation between lattice-valued Kripke structures (LKSs), and establish the correspondence between lattice-valued bisimulation in LLTS and lattice-valued bisimulation in LKS.

Many researchers suggested extending bisimilarity to quantitative versions to avoid the rigidity of classical bisimilarity. To explore the relation between different notions of approximate bisimilarity mentioned in literature, in this paper, we present a quantitative extension of doubly labeled transition systems, MDLTS, where its states and actions form metric spaces. We then introduce two notions of approximate bisimilarity, (η, λ)-bisimilarity and (η, λ, α)-bisimilarity, and discuss their basic property. We also consider the special kind of (η, λ)-bisimilarity, λ-bisimilarity to characterize the branching distance with arbitrary discount α of metric labeled transition system. Finally, we discuss the translation between metric transition system and MDLTS which preserves the approximate bisimilarity.

In this paper, we focus on the infinite evolution of the parameterized bisimulation in order to discuss the dynamic characterization of programs. We propose parameterized limit bisimulation and parameterized bisimulation limit which are useful for understanding and analyzing of infinite evolution of concurrent programs. Some special parameterized limit bisimulations are introduced and some topological properties are proved.

In this paper, we give a bigraphical model for web services composition. We investigate how to represent scope-based compensation handing mechanism by means of Bigraphical Reactive Systems (13RSs for short), which have been proposed to provide a uniform way to model spatially distributed systems that both compute and communicate. The service composition language we focus on is WSBPEL, which is the standard of web service composition and orchestration. This bigraphical model can be regarded as a unifying semantics of BPEL-like languages with the key concepts related to compensation handling. The rationality of the model is discussed by investigating the relationship between BPEL language and BRSs. Based on the bigraphical model, the algebraic laws for BPEL are proved as well.