Recently,Generative Adversarial Networks(GANs)have become the mainstream text-to-image(T2I)***,a standard normal distribution noise of inputs cannot provide sufficient information to synthesize an image that approache...
详细信息
Recently,Generative Adversarial Networks(GANs)have become the mainstream text-to-image(T2I)***,a standard normal distribution noise of inputs cannot provide sufficient information to synthesize an image that approaches the ground-truth image ***,the multistage generation strategy results in complex T2I ***,this study proposes a novel feature-grounded single-stage T2I model,which considers the“real”distribution learned from training images as one input and introduces a worst-case-optimized similarity measure into the loss function to enhance the model's generation *** results on two benchmark datasets demonstrate the competitive performance of the proposed model in terms of the Frechet inception distance and inception score compared to those of some classical and state-of-the-art models,showing the improved similarities among the generated image,text,and ground truth.
Cardiovascular disease (CVD) is a major contributor to the current rise in mortality rates, so accurate prediction of CVD is crucial. Traditional prediction methods often rely on a single type of data, ignoring the co...
详细信息
Offline reinforcement learning(RL) has gathered increasing attention in recent years, which seeks to learn policies from static datasets without active online exploration. However, the existing offline RL approaches o...
详细信息
Offline reinforcement learning(RL) has gathered increasing attention in recent years, which seeks to learn policies from static datasets without active online exploration. However, the existing offline RL approaches often require a large amount of pre-collected data and hence are hardly implemented by a single agent in practice. Inspired by the advancement of federated learning(FL), this paper studies federated offline reinforcement learning(FORL),whereby multiple agents collaboratively carry out offline policy learning with no need to share their raw ***, a straightforward solution is to simply retrofit the off-the-shelf offline RL methods for FL, whereas such an approach easily overfits individual datasets during local updating, leading to instability and subpar performance. To overcome this challenge, we propose a new FORL algorithm, named model-free(MF)-FORL, that exploits novel“proximal local policy evaluation” to judiciously push up action values beyond local data support, enabling agents to capture the individual information without forgetting the aggregated knowledge. Further, we introduce a model-based variant, MB-FORL, capable of improving the generalization ability and computational efficiency via utilizing a learned dynamics model. We evaluate the proposed algorithms on a suite of complex and high-dimensional offline RL benchmarks, and the results demonstrate significant performance gains over the baselines.
Temporal knowledge graph(TKG) reasoning, has seen widespread use for modeling real-world events, particularly in extrapolation settings. Nevertheless, most previous studies are embedded models, which require both enti...
详细信息
Temporal knowledge graph(TKG) reasoning, has seen widespread use for modeling real-world events, particularly in extrapolation settings. Nevertheless, most previous studies are embedded models, which require both entity and relation embedding to make predictions, ignoring the semantic correlations among different entities and relations within the same timestamp. This can lead to random and nonsensical predictions when unseen entities or relations occur. Furthermore, many existing models exhibit limitations in handling highly correlated historical facts with extensive temporal depth. They often either overlook such facts or overly accentuate the relationships between recurring past occurrences and their current counterparts. Due to the dynamic nature of TKG, effectively capturing the evolving semantics between different timestamps can be *** address these shortcomings, we propose the recurrent semantic evidenceaware graph neural network(RE-SEGNN), a novel graph neural network that can learn the semantics of entities and relations simultaneously. For the former challenge, our model can predict a possible answer to missing quadruples based on semantics when facing unseen entities or relations. For the latter problem, based on an obvious established force, both the recency and frequency of semantic history tend to confer a higher reference value for the current. We use the Hawkes process to compute the semantic trend, which allows the semantics of recent facts to gain more attention than those of distant facts. Experimental results show that RE-SEGNN outperforms all SOTA models in entity prediction on 6 widely used datasets, and 5 datasets in relation prediction. Furthermore, the case study shows how our model can deal with unseen entities and relations.
Oxygen evolution reaction(OER)is one of the most important half-reactions related to metal-air batteries,fuel cells,and *** to the sluggish kinetic and multi-electron transfer,catalysts appear to be particularly impor...
详细信息
Oxygen evolution reaction(OER)is one of the most important half-reactions related to metal-air batteries,fuel cells,and *** to the sluggish kinetic and multi-electron transfer,catalysts appear to be particularly important for the *** the reaction mechanism is fundamental to developing new catalysts and improving OER *** this work,phase transition and atomic reconstruction on CoO(111)plane were revealed through ex-situ diffraction methods and X-ray absorption *** the same time,the electronic state evolution of Co(II)/Co(III)during the OER process has also been concluded by analyzing the magnetic *** work shows that during the OER process,Co(III)experiences surface electron rearrangement from IS(intermediate-spin state)to LS(low-spin state)and then returns to IS/HS(high-spin state)under high voltage *** work provides a new view to reveal the reaction mechanism through the magnetic property and it can be extended to more magnetic 3d transition metals for future catalyst design.
Accurate localization of tropical cyclone (TC) centers is crucial for intensity estimation and track prediction. While traditional methodologies primarily rely on visible light and infrared (IR) imagery, the potential...
详细信息
Simulating the total ionizing dose(TID)of an electrical system using transistor-level models can be difficult and expensive,particularly for digital-integrated circuits(ICs).In this study,a method for modeling TID eff...
详细信息
Simulating the total ionizing dose(TID)of an electrical system using transistor-level models can be difficult and expensive,particularly for digital-integrated circuits(ICs).In this study,a method for modeling TID effects in complementary metaloxide semiconductor(CMOS)digital ICs based on the input/output buffer information specification(IBIS)was *** digital IC was first divided into three parts based on its internal structure:the input buffer,output buffer,and functional *** of these three parts was separately *** the IBIS model,the transistor V-I characteristic curves of the buffers were processed,and the physical parameters were extracted and modeled using *** the functional area,logic functions were modeled in VHDL according to the data sheet.A golden digital IC model was developed by combining the input buffer,output buffer,and functional area ***,the golden ratio was reconstructed based on TID experimental data,enabling the assessment of TID effects on the threshold voltage,carrier mobility,and time series of the digital *** experiments were conducted using a CMOS non-inverting multiplexer,NC7SZ157,and the results were compared with the simulation results,which showed that the relative errors were less than 2%at each dose *** confirms the practicality and accuracy of the proposed modeling *** TID effect model for digital ICs developed using this modeling technique includes both the logical function of the IC and changes in electrical properties and functional degradation impacted by TID,which has potential applications in the design of radiation-hardening tolerance in digital ICs.
The discourse analysis task,which focuses on understanding the semantics of long text spans,has received increasing attention in recent *** a critical component of discourse analysis,discourse relation recognition aim...
详细信息
The discourse analysis task,which focuses on understanding the semantics of long text spans,has received increasing attention in recent *** a critical component of discourse analysis,discourse relation recognition aims to identify the rhetorical relations between adjacent discourse units(e.g.,clauses,sentences,and sentence groups),called arguments,in a *** works focused on capturing the semantic interactions between arguments to recognize their discourse relations,ignoring important textual information in the surrounding ***,in many cases,more than capturing semantic interactions from the texts of the two arguments are needed to identify their rhetorical relations,requiring mining more contextual *** this paper,we propose a method to convert the RST-style discourse trees in the training set into dependency-based trees and train a contextual evidence selector on these transformed *** this way,the selector can learn the ability to automatically pick critical textual information from the context(i.e.,as evidence)for arguments to assist in discriminating their *** we encode the arguments concatenated with corresponding evidence to obtain the enhanced argument ***,we combine original and enhanced argument representations to recognize their *** addition,we introduce auxiliary tasks to guide the training of the evidence selector to strengthen its selection *** experimental results on the Chinese CDTB dataset show that our method outperforms several state-of-the-art baselines in both micro and macro F1 scores.
Author name disambiguation(AND)is a central task in academic search,which has received more attention recently accompanied by the increase of authors and academic *** tackle the AND problem,existing studies have propo...
详细信息
Author name disambiguation(AND)is a central task in academic search,which has received more attention recently accompanied by the increase of authors and academic *** tackle the AND problem,existing studies have proposed various approaches based on different types of information,such as raw document features(e.g.,co-authors,titles,and keywords),the fusion feature(e.g.,a hybrid publication embedding based on multiple raw document features),the local structural information(e.g.,a publication's neighborhood information on a graph),and the global structural information(e.g.,interactive information between a node and others on a graph).However,there has been no work taking all the above-mentioned information into account and taking full advantage of the contributions of each raw document feature for the AND problem so *** fill the gap,we propose a novel framework named EAND(Towards Effective Author Name Disambiguation by Hybrid Attention).Specifically,we design a novel feature extraction model,which consists of three hybrid attention mechanism layers,to extract key information from the global structural information and the local structural information that are generated from six similarity graphs constructed based on different similarity coefficients,raw document features,and the fusion *** hybrid attention mechanism layer contains three key modules:a local structural perception,a global structural perception,and a feature ***,the mean absolute error function in the joint loss function is used to introduce the structural information loss of the vector *** results on two real-world datasets demonstrate that EAND achieves superior performance,outperforming state-of-the-art methods by at least+2.74%in terms of the micro-F1 score and+3.31%in terms of the macro-F1 score.
ExpertRecommendation(ER)aims to identify domain experts with high expertise and willingness to provide answers to questions in Community Question Answering(CQA)web *** to model questions and users in the heterogeneous...
详细信息
ExpertRecommendation(ER)aims to identify domain experts with high expertise and willingness to provide answers to questions in Community Question Answering(CQA)web *** to model questions and users in the heterogeneous content network is critical to this *** traditional methods focus on modeling questions and users based on the textual content left in the community while ignoring the structural properties of heterogeneous CQA networks and always suffering from textual data sparsity *** approaches take advantage of structural proximities between nodes and attempt to fuse the textual content of nodes for ***,they often fail to distinguish the nodes’personalized preferences and only consider the textual content of a part of the nodes in network embedding learning,while ignoring the semantic relevance of *** this paper,we propose a novel framework that jointly considers the structural proximity relations and textual semantic relevance to model users and questions more ***,we learn topology-based embeddings through a hierarchical attentive network learning strategy,in which the proximity information and the personalized preference of nodes are encoded and ***,we utilize the node’s textual content and the text correlation between adjacent nodes to build the content-based embedding through a meta-context-aware skip-gram *** addition,the user’s relative answer quality is incorporated to promote the ranking *** results show that our proposed framework consistently and significantly outperforms the state-of-the-art baselines on three real-world datasets by taking the deep semantic understanding and structural feature learning *** performance of the proposed work is analyzed in terms of MRR,P@K,and MAP and is proven to be more advanced than the existing methodologies.
暂无评论