Distant supervision is an efficient way to generate large-scale training data for relation extraction without human efforts. However, the accompanying challenges have been plaguing the advance of the extractor: (1) the automatically annotated labels for training data contain much noisy data; (2) the annotations, based on bag-level (cluster of sentences) instead of sentence-level (single sentence), are too coarse to train an accurate extractor; (3) hetergeneous sentences are hard for a denoising model to capture the underlying commonness among valid relational expressions. To address these issues, we bulid a novel sentence representation and craft reinforcement learning to select the expressive sentence for each relation mentioned in a bag. More specifically, we introduce entity-free sentence pattern incorporated with attentive type information. Furthermore, multiple interactions between entity-specific and entity-free representation are proposed to generate complementary sentence features (for challenge 3). Then we design a fine-grained reward function, and model the sentence selection process as an auction where different relations for a bag need to compete together to achieve the possession of a specific sentence based on its expressiveness(for challenge 1 and 2). The experimental results on two public datasets demonstrate the superiority of our model for distantly supervised relation extraction.

Model-based diagnosis (MBD) with multiple observations is complicated and difficult to manage over. In this paper, we proposed two new diagnosis models, namely, the Compacted Model with Multiple Observations (CMMO) and the Dominated-based Compacted Model with Multiple Observations (D-CMMO), to solve the problem in which a considerable amount of time is needed when multiple observations are given and more than one fault is injected. Three ideas are presented in this paper. First, we propose to encode MBD with each observation as a subsystem and share as many system variables as possible to compress the size of encoded clauses. Second, we utilize the notion of gate dominance in the CMMO approach to compute Top-Level Diagnosis with Compacted Model (CM-TLD) to reduce the solution space. Finally, we explore the performance of our model using three fault models. Experimental results on the ISCAS-85 benchmarks show that CMMO and D-CMMO perform better than the state-of-the-art algorithms.

For many high-tech fields such as space exploration, nuclear technology, and smart automobiles, it is vital to timely find faulty components of man-made devices to ensure safety. However, there is nearlynoenough diagnostic experience accumulated in these new devices, and thus, it is hardly suitable to only apply the traditional expert/experience-based fault diagnosis approach. Thus, model-based diagnosis was proposed for efficient detection of faulty components; this approach explores the behavioral and structural information of the device to be diagnosed, and no experience is required. In model-based diagnosis, for a device to be diagnosed, minimal conflict sets of components are first generated, and all minimal hitting-sets for them will be derived as candidate diagnoses. Therefore, it is vital to efficiently generate all minimal hitting-sets to find the final diagnosis. Unfortunately, it is proven to be NP-hard when deriving all minimal hitting-sets for given minimal conflict sets. To improve the computing efficiency, in this article, we propose a novel approach calledTreeMerge, which considers a special type oftreestructure of minimal conflict sets of large sizes since structural information usually plays an important role in solving complex problems. Theoretically, compared with other algorithms, the time complexity of the new algorithm is greatly reduced, as the time complexity of the new algorithm becomeslinearrather thanquadratic. Furthermore, experimental results on multiple synthetic and benchmark examples show that the proposedTreeMergealgorithm is more efficient than many other state-of-the-art methods, with a reduction ofseveral orders of magnituderuntime (seconds).

Planning and model-based diagnosis are both branches of artificial intelligence. In model-based diagnosis, multiresults may be gotten which lead to an uncertain diagnosis. We use the landmark method from planning to designing an event sequence to get a reaction. A method that uses planning to repair in local results of incremental diagnosis is proposed. Firstly, a model is established on model-based diagnosis and planning. Incremental diagnosis results are used as the initial state of planning, and the heuristic search method is used to find the solution to an unfaulty state. Two algorithms with different strategies are designed for diagnosis and repair: one is to repair all possible faults and use controllable events to repair them, and the other is to test through the feedback of controllable events and observable events to get the only solution and repair them. At the same time, the efficiency of the incremental diagnosis method is improved based on heuristics.

Distant supervision is an efficient way to generate large-scale training data for relation extraction without human efforts. However, a coin has two sides. The automatically annotated labels for training data are problematic, which can be summarized as multi-instance multi-label problem and coarse-grained (bag-level) supervised signal. To address these problems, we propose two reasonable assumptions and craft reinforcement learning to capture the expressive sentence for each relation mentioned in a bag. More specifically, we extend the original expressed-at-least-once assumption to multi-label level, and introduce a novel express-at-most-one assumption. Besides, we design a fine-grained reward function, and model the sentence selection process as an auction where different relations for a bag need to compete together to achieve the possession of a specific sentence based on its expressiveness. In this way, our model can be dynamically self-adapted, and eventually implements the accurate one-to-one mapping from a relation label to its chosen expressive sentence, which serves as training instances for the extractor. The experimental results on a public dataset demonstrate that our model constantly and substantially outperforms current state-of-the-art methods for relation extraction.

Finding justifications for an entailment is one of the major missions in the field of ontology research. Recent advances on finding justifications w.r.t. the light-weight description logics focused on encoding this problem into a propositional formula, and using SAT-based techniques to enumerate all MUSes (minimally unsatisfiable subformulas). It's necessary to import more optimized techniques into finding justifications as emergence of large-scale real-world ontologies. In this paper, we propose a new strategy which introduce local search(in short, LS) technique to compute the approximating core before extracting an exact MUS. Although it is based on a heuristic and LS, such technique is complete in the sense that it always delivers a MUS for any unsatisfiable SAT instance. Our method will find the justifications for large-scale ontologies more effectively.

The recent art in relation extraction is distant supervision which generates training data by heuristically aligning a knowledge base with free texts and thus avoids human labelling. However, the concerned relation mentions often use the bag-of-words representation, which ignores inner correlations between features located in different dimensions and makes relation extraction less effective. To capture the complex characteristics of relation expression and tighten the correlated features, we attempt to discover and utilise informative correlations between features by the following four phases: 1) formulating semantic similarities between lexical features using the embedding method; 2) constructing generative relation for lexical features with different sizes of side windows; 3) computing correlation scores between syntactic features through a kernel-based method; and 4) conducting a distillation process for the obtained correlated feature pairs and integrating informative pairs with existing relation extraction models. The extensive experiments demonstrate that our method can effectively discover correlation information and improve the performance of state-of-the-art relation extraction methods.

Deriving all minimal hitting sets (MHSes) for a family of conflict sets is a classical problem in model-based diagnosis. A technique for distributed MHSes based on the join relation of elements is proposed. Then, a strategy for deriving all distributed MHSes is presented. If the family of sets is decomposed into a number of equivalence classes based on the join relation, then parallel computation of MHSes for each distribution can be applied. Moreover, an incremental, distributed approach is introduced. When a new conflict set is added, only related distributed MHSes are chosen to incrementally update the final result. From a theoretical point of view, the complexity of the distributed algorithm is O(2num/k), while the complexity of the corresponding centralized algorithm is O(2num), with k and num being the number of equivalence classes and the number of basic elements in all the conflict sets, respectively. Furthermore, compared with the corresponding centralized approach, a large number of set-containment checks are avoided by the incremental, distributed approach. Experimental results, including both numerous artificial examples and typical International Symposium on Circuits and Systems-85 benchmark circuit conflict set examples, offer evidence that, compared with centralized methods, the efficiency for deriving all MHSes in a distributed (incremental) way is considerably improved.

This paper addresses the problem of failure diagnosis in component-based discrete event systems. In this paper we propose a method to obtain the set of components when dealing with diagnosis in large complex discrete event systems. In the new method, before disassembling the system into components, we need to identify whether insert communication events into the system or not. When analyzing the diagnosability, we treat the system containing communication events as a distributed discrete event system. Otherwise we treat the system as a decentralized discrete event system. For the components which are not diagnosable, we propose a method to reconstruct them by utilizing some other components sharing the same communication events with them. This algorithm provides more accurate information of the diagnosability of the system.

Model-based diagnosis of discrete event systems (DESs) is more and more active in artificial intelligence. However, there has been always a very restrictive assumption in the previous works that the model of a given DES is complete, including all nominal behaviors and all possible failure behaviors of the system. In order to relax this so restrictive assumption, in this paper, model-based diagnosis of a DES with an incomplete system model is investigated. A new concept of “P-synchronization product” of finite state automata is proposed, by which the P-diagnosis of the DES with an incomplete system model is easily put forward. It is also shown that the traditional synchronization product of finite state automata can be seen as a special situation of P-synchronization product. In addition, an ideal heuristic way from theoretical view to improve the P-synchronization product is discussed as well.

Model-based diagnosis is an effective approach without needing expert experience, whose one key step is to identify the conflict sets. According to General Diagnostic Engine (GDE), an approach to identify the minimal conflict sets based on Assumption-based Truth Maintenance System (ATMS) is proposed. By defining negative-node and some new rules, the incompleteness of ATMS is overcome. This approach is incremental through propagating environments. It is sound, complete, consistent, and all of solutions are minimal. Moreover, simulations show that it is effective, general and easy to be implemented and expanded as well.

Model-based diagnosis of discrete event systems is more and more active in artificial intelligence. In this paper, diagnosability analysis of discrete event systems is concerned, which is a very important step before on line diagnosing discrete event systems in general. Firstly, an extended hierarchical framework for definitions of diagnosability of discrete event systems is given, according to their inner restriction. Next, some formal comparisons among them are presented, thanks to which, we can further understand the relations between related definitions. Finally, some future work about diagnosability of discrete event systems is discussed as well.