ABSTRACT Relation extraction, as a crucial component of knowledge engineering, is essential to the automated construction of knowledge‐based expert systems. The existing methods ignore the important phrase information, and other useful information in the sentence is not fully utilised. To address these problems, we propose the phrase information enhanced and relation guided relation extraction, named PIERG. This method uses a multi‐scale convolutional neural network, which is sensitive to local context, to obtain multi‐granularity phrase features, and applies the gating mechanism to filter out the noise information in a sentence and find out the key phrase features that trigger relation types. In addition, an attentional network enhanced by entity pairs and piecewise information is designed to learn more useful information related to relation types. Finally, by calculating the correlation score between the decomposed relation labels and the head entity, the tail entity and the final sentence vector representation, the relation types can be more effectively located. The experiments demonstrate that PIERG can achieve better extraction results than most previous methods. Therefore, PIERG can serve as an effective technical pathway for automated knowledge acquisition and knowledge base filling in expert systems, substantially reducing manual effort in structured knowledge engineering.

Recent advances in differentiable score-based methods for Directed Acyclic Graph (DAG) structure learning have revolutionized the problem of combinatorial structure learning, transforming it into a continuous optimization task. Despite their remarkable success, these methods rely on a key assumption that all samples have the same level of difficulty and no data heterogeneity. When this assumption does not hold, causal discovery algorithms based on it inevitably return networks with many spurious edges. Despite existing research, the current method ignores the reality of outliers in the samples, introducing certain limitations that still result in erroneous edges. Inspired by the rapid decay of the Gaussian distribution as distance from the center increases, we propose an innovative adaptive sample reweighting framework based on asymmetric exponential modulation Gaussian, coined DAG-AEG. DAG-AEG boosts DAG structure learning by analyzing the distribution of sample losses and employing the proposed method for adaptive sample attention. Additionally, it can be adapted to heterogeneous data. We used various causal structure learning methods to test the performance of DAG-AEG on synthetic and real datasets. The experimental results demonstrate that the proposed framework significantly improves the performance across all methods, outperforming existing methods.

Meta-reinforcement learning with contexts enables the policy to quickly adapt to new tasks, which is important for bridging the gap between the simulation to the real uncertain environment. However, the learning of traditional context encoders relies on transition data collected during meta-training, which results in all environments being mapped into a single encoding space and reduces the sensitivity of context encoders to environmental uncertainties. This work offers the Robust Meta Reinforcement Learning via Environment Context Enhancement (RMRL-CE) algorithm to enhance adaptability to abrupt changes in uncertain environments and to augment the robustness of the policy. Firstly, we use the local-global context encoders which offer more precise environmental data for decision-making. Subsequently, we train the global context encoder by intra-class loss to simplify the segmentation of the context embedding space, which improves the context encoder’s sensitivity to environmental uncertainty. Finally, the comprehensive meta-policy is developed alongside the conditional value-at-risk (CVaR) optimization method. We used a set of sequential control tasks on MuJoCo to make meta-test tasks. Our method outperformed the benchmark methods on both mean and CVaR returns, demonstrating high robustness even under harsh uncertain environmental conditions.

Deploying reinforcement learning into real-world environments remains challenging at present due to the impact of environmental uncertainty. Enhancing the generalization ability of agents can help improve this problem. Curriculum learning provides an effective means to improve model generalization by simulating the teacher-student training mode. The teacher module generates progressively more difficult training tasks, enabling the student module to learn task skills more efficiently than with disorganized tasks. However, it is not easy to define the difficulty of the curriculum, and not stable enough for training if only relying on student reward feedback in curriculum learning. In this paper, we argue that the model prediction uncertainty of agents more accurately reflects its environmental adaptation level and can thus reliably determine curriculum difficulty. Thus, we propose Uncertainty-Guided Curriculum Design (UGCD), which aims to design training tasks that align with the frontier of agent ability, so that agents can improve their generalization ability in more challenging environments. First, we evaluate the value prediction uncertainty of agents through the ensemble probabilistic neural networks and use it to define the difficulty of the curriculum. Moreover, we utilize the novelty score to ensure diversity in environment generation. A curriculum that combines both challenge and novelty will prevent the agents from getting stuck in learning dilemmas and enable them to learn task skills gradually. Our approach achieves better performance on a series of zero-shot tasks, which suggests that uncertainty-guided lessons are effective in improving the generalization ability of agents.

Abstract SpMV (Sparse matrix‐vector multiplication) is an important computing core in traditional high‐performance computing and also one of the emerging data‐intensive applications. For diagonal sparse matrices, it is frequently necessary to fill in a large number of zeros to maintain the diagonal structure as for using DIA (Diagonal) storage format. The fact that filling with zeros may consume additional computing and memory resources, will certainly lead to degradation of the parallel computing performance of SpMV, further causing computing and storage redundancy. To solve the deficiencies of the DIA format, a Two‐stage parallel SpMV method is presented in this paper, which can reasonably distribute the data of diagonal matrix and irregular matrix to different CUDA kernels. As different corresponding compression methods are particularly designed for different matrix forms, a partition‐based hybrid format of DIA and CSR (HPDC) is therefore adopted in the two‐stage method to ensure load balancing among computing resources and continuity of data access on the diagonal. Simultaneously, a standard deviation among blocks is used as a criterion to obtain the optimal number of blocks and distribution of data. The experimental data were implemented in the Florida data set. Compared to DIA, cuSPARSE‐CSR, HDC, and BRCSD, the execution time of the Two‐stage method is shortened by 4, 3.4, 1.9, and 1.15, respectively.

Participatory sensing is often used in environmental or personal data monitoring, wherein a number of participants collect data using their mobile intelligent devices for earning the incentives. However, a lot of additional information is submitted along with the data, such as the participant’s location, IP and incentives. This multimodal information implicitly links to the participant’s identity and exposes the participant’s privacy. In order to solve the issue of these multimodal information associating with participants’ identities, this paper proposes a protocol to ensure anonymous data reporting while providing a dynamic incentive mechanism simultaneously. The proposed protocol first establishes a submission schedule by anonymously selecting a slot in a vector by each member where every member and system entities are oblivious of other members’ slots and then uses this schedule to submit the all members’ data in an encoded vector through bulk transfer and multiplayer dining cryptographers networks (DC-nets) . Hence, the link between the data and the member’s identity is broken. The incentive mechanism uses blind signature to anonymously mark the price and complete the micropayments transfer. Finally, the theoretical analysis of the protocol proves the anonymity, integrity, and efficiency of this protocol. We implemented and tested the protocol on Android phones. The experiment results show that the protocol is efficient for low latency tolerable applications, which is the cases with most participatory sensing applications, and they also show the advantage of our optimization over similar anonymous data reporting protocols.

Distant supervision (DS) has been widely used for relation extraction (RE), which automatically generates large‐scale labeled data. However, there is a wrong labeling problem, which affects the performance of RE. Besides, the existing method suffers from the lack of useful semantic features for some positive training instances. To address the above problems, we propose a novel RE model with sentence selection and interaction representation for distantly supervised RE. First, we propose a pattern method based on the relation trigger words as a sentence selector to filter out noisy sentences to alleviate the wrong labeling problem. After clean instances are obtained, we propose the interaction representation using the word‐level attention mechanism‐based entity pairs to dynamically increase the weights of the words related to entity pairs, which can provide more useful semantic information for relation prediction. The proposed model outperforms the strongest baseline by 2.61 in F1‐score on a widely used dataset, which proves that our model performs significantly better than the state‐of‐the‐art RE systems.