A major trend in self-driving technology is to introduce new causal reasoning mechanisms for decision-making. This paper mainly discusses the problem of causal direction reasoning for nonlinear noisy data. The paper presents the errors-in-variables (EIV) system to construct a causality model and the Hilbert-Schmidt independence criterion (HSIC) to compute the dependence between variables. Then, the paper proposes a new method that is based on the EIV model and HSIC. In the proposed method, noise at both the input and output is considered simultaneously. The proposed method has strong robustness and maintains a relatively stable inference accuracy when the observational noise is considerable. Experiments on simulated and real-world data are presented to demonstrate the performance of the proposed method.

One of recent trends [31, 32, 14] in network architecture design is stacking small filters (e.g., 1×1 or 3×3) in the entire network because the stacked small filters is more efficient than a large kernel, given the same computational complexity. However, in the field of semantic segmentation, where we need to perform dense per-pixel prediction, we find that the large kernel (and effective receptive field) plays an important role when we have to perform the classification and localization tasks simultaneously. Following our design principle, we propose a Global Convolutional Network to address both the classification and localization issues for the semantic segmentation. We also suggest a residual-based boundary refinement to further refine the object boundaries. Our approach achieves state-of-art performance on two public benchmarks and significantly outperforms previous results, 82.2% (vs 80.2%) on PASCAL VOC 2012 dataset and 76.9% (vs 71.8%) on Cityscapes dataset.

The principle of counter-transitivity plays a vital role in argumentation. It states that an argument is strong when its attackers are weak, and is weak when its attackers are strong. In this work, we develop a formal theory about the argument ranking semantics based on this principle. Three approaches, quantity-based, quality-based and the unity of them, are defined to implement the principle. Then, we show an iterative refinement algorithm for capturing the ranking on arguments based on the recursive nature of the principle.

Dung's abstract argumentation framework consists of a set of interacting arguments and a series of semantics for evaluating them. Those semantics partition the powerset of the set of arguments into two classes: extensions and non-extensions. In order to reason with a specific semantics, one needs to take a credulous or skeptical approach, i.e. an argument is eventually accepted, if it is accepted in one or all extensions, respectively. In our previous work [1], we have proposed a novel semantics, called counting semantics, which allows for a more fine-grained assessment to arguments by counting the number of their respective attackers and defenders based on argument graph and argument game. In this paper, we continue our previous work by presenting some supplementaries about how to choose the damaging factor for the counting semantics, and what relationships with some existing approaches, such as Dung's classical semantics, generic gradual valuations. Lastly, an axiomatic perspective on the ranking semantics induced by our counting semantics are presented.

The knowledgebase uncertainty and the argument preferences are considered in this paper. The uncertainty is captured by weighted satisfiability degree, while a preference relation over arguments is derived by the beliefs of an agent.

Algorithm's time complexity is an essential issue for time series prediction in numerous practices.A novel fast exact inference method for Gaussian process model is proposed in this paper to accelerate the task of non-stationary time series prediction. Experiment was done on the real world power load data.

Inferring the causal direction between two variables is a nontrivial problem in the subject of causal discovery from observed data. A method for errors-in-variables models where both the cause variable and the effect variable are observed with measurement errors is presented in this paper.

Compositional verification of invariants is a technique for alleviating the state explosion problem in component-based verification. The efficiency of these methods depends on abstraction, which leads to verification incompleteness. In this paper, we present a unified framework that combines compositional abstraction and counterexample-guided abstraction refinement (CEGAR) to address this incompleteness problem. We propose two refinement approaches: invariant strengthening and state partitioning. In the case of a spurious counterexample, our proposed invariant strengthening approach refines the abstraction by eliminating the infeasible states. The state partitioning approach exploits the semantics of component based systems and obtains a more precise system. Any safety property that holds on the abstraction is guaranteed to hold on the model refined by the state partitioning approach. The examples and experiments in this paper show that our verification method can achieve conclusive results in the verification of safety properties with deadlock freedom in component-based systems.

Model checking has had a substantive impact on the verification of software, and nowadays more and more software are developed with components manufactured by a third party. This paper examines the verification of systems with unspecified components. The method is based on the model of oracle automaton and related checking algorithms. Two problems in the previous checking algorithm are found and analyzed. To correct the previous error, an efficient algorithm for verifying the system with unspecified components is introduced. The new algorithm not only corrects the errors but also extends the original algorithm to multi-oracle finite automaton. The set of input transitions that are not related to the oracles is minimized by the new algorithm. Further, a prototype tool for checking oracle finite automaton (OFA) is implemented. It has advantages for modeling systems with unspecified components and merging different checking algorithms. Finally, the architecture and design of this tool are illustrated.