16,063 Hits in 6.2 sec

Interpreting and Unifying Graph Neural Networks with An Optimization Framework [article]

Meiqi Zhu, Xiao Wang, Chuan Shi, Houye Ji, Peng Cui
2021 arXiv   pre-print
the feasibility for designing GNNs with our unified optimization framework.  ...  Graph Neural Networks (GNNs) have received considerable attention on graph-structured data learning for a wide variety of tasks.  ...  This further verifies the feasibility for designing GNNs under the unified framework. RELATED WORK Graph Neural Networks.  ... 
arXiv:2101.11859v1 fatcat:45ffciqonrgjxe7dcn7546oksi

Unifying Neural Learning and Symbolic Reasoning for Spinal Medical Report Generation [article]

Zhongyi Han, Benzheng Wei, Yilong Yin, Shuo Li
2020 arXiv   pre-print
Concretely, we design an adversarial graph network that interpolates a symbolic graph reasoning module into a generative adversarial network through embedding prior domain knowledge, achieving semantic  ...  In this paper, we propose the neural-symbolic learning (NSL) framework that performs human-like learning by unifying deep neural learning and symbolic logical reasoning for the spinal medical report generation  ...  The learning strategy of the adversarial graph network has two stages: 1) construct the symbolic graph, and 2) optimize the network.  ... 
arXiv:2004.13577v1 fatcat:oh5aka5zr5be3ipd7qqnyhikzy

Neuro-Symbolic Learning: Principles and Applications in Ophthalmology [article]

Muhammad Hassan, Haifei Guan, Aikaterini Melliou, Yuqi Wang, Qianhui Sun, Sen Zeng, Wen Liang, Yiwei Zhang, Ziheng Zhang, Qiuyue Hu, Yang Liu, Shunkai Shi (+15 others)
2022 arXiv   pre-print
Neural networks have been rapidly expanding in recent years, with novel strategies and applications.  ...  However, challenges such as interpretability, explainability, robustness, safety, trust, and sensibility remain unsolved in neural network technologies, despite the fact that they will unavoidably be addressed  ...  Graph neural networks (GNNs) generally learn an embed-ding that maps nodes and edges into a continuous vector space with limitations on the number of vertices in the input graph [77] .  ... 
arXiv:2208.00374v1 fatcat:pktmnomj3bbwpjyj7lmu37rl7i

Book announcements

1993 Discrete Applied Mathematics  
Ramacher and R. Ruckert, eds., VLSI Design of Neural Networks (Kluwer Academic Publishers, Boston, 1991) 343 pages An entry in the middle of an iteration. Non-unique exit in an iteration).  ...  An Initial Algebra Framework for Unifying the Structured Models. Introduction. Algebras. Initial algebras. Yourdon structure charts. DeMarco data flow diagrams. Jackson structure texts.  ...  Example on an application  ... 
doi:10.1016/0166-218x(93)90117-7 fatcat:mo4kjmd2czdg5hejcof7fmm6u4

Performance Comparison between Pytorch and Mindspore

Xiangyu XIA, Shaoxiang ZHOU
2022 International Journal of Database Management Systems  
However, there is a large amount of data when training neural networks, which makes many deep learning frameworks appear to serve deep learning practitioners, providing services that are more convenient  ...  To ensure that our survey is as comprehensive as possible, we carefully selected neural networks in 2 main domains, which cover computer vision and natural language processing (NLP).  ...  ACKNOWLEDGEMENTS Supported by Beijing City University in 2021 "the innovation and entrepreneurship training program for college students"  ... 
doi:10.5121/ijdms.2022.14201 fatcat:vpq5wvjfnfbl7juh6i5572z6ze


V. N. Pakhomova, I. D. Tsykalo
2019 Nauka ta Progres Transportu  
The use of a neural model, the input of which is an array of channel bandwidth, will allow in real time to determine the optimal route in the computer network.  ...  It allows to perform the following steps: sample generation (random or balanced); creation of a neural network, the input of which is an array of bandwidth of the computer network channels; training and  ...  Python is an interpreted object-oriented highlevel programming language with strict dynamic typing.  ... 
doi:10.15802/stp2018/154443 fatcat:l5yokcdx2rdjna43k2aocvswm4

Generative Causal Explanations for Graph Neural Networks [article]

Wanyu Lin and Hao Lan and Baochun Li
2021 arXiv   pre-print
Our theoretical analysis shows that several recent explainers fall into a unified framework of additive feature attribution methods.  ...  This paper presents Gem, a model-agnostic approach for providing interpretable explanations for any GNNs on various graph learning tasks.  ...  We show that causal interpretability could contribute to explaining and understanding graph neural networks.  ... 
arXiv:2104.06643v2 fatcat:j3yopmmcrffyvc4mklmdh6tr2y

GmCN: Graph Mask Convolutional Network [article]

Bo Jiang, Beibei Wang, Jin Tang, Bin Luo
2019 arXiv   pre-print
Graph Convolutional Networks (GCNs) have shown very powerful for graph data representation and learning tasks.  ...  GmCN can be theoretically interpreted by a regularization framework, based on which we derive a simple update algorithm to determine the optimal mask adaptively in GmCN training process.  ...  A new unified regularization framework is derived for GmCN interpretation, based on which we provide an effective update algorithm to achieve GmCN optimization.  ... 
arXiv:1910.01735v2 fatcat:mzgmlpc5nrccbgqbmxktnecw7i

Graph Optimal Transport for Cross-Domain Alignment [article]

Liqun Chen, Zhe Gan, Yu Cheng, Linjie Li, Lawrence Carin, Jingjing Liu
2020 arXiv   pre-print
We propose Graph Optimal Transport (GOT), a principled framework that germinates from recent advances in Optimal Transport (OT).  ...  Both WD and GWD can be incorporated into existing neural network models, effectively acting as a drop-in regularizer.  ...  The research at Duke University was supported in part by DARPA, DOE, NIH, NSF and ONR.  ... 
arXiv:2006.14744v3 fatcat:scfmjoxrsbcydcey4r5pfvejdu

A Unified and Biologically-Plausible Relational Graph Representation of Vision Transformers [article]

Yuzhong Chen, Yu Du, Zhenxiang Xiao, Lin Zhao, Lu Zhang, David Weizhong Liu, Dajiang Zhu, Tuo Zhang, Xintao Hu, Tianming Liu, Xi Jiang
2022 arXiv   pre-print
Overall, our work provides a novel unified and biologically-plausible paradigm for more interpretable and effective representation of ViT ANNs.  ...  Using this unified relational graph representation, we found that: a) a sweet spot of the aggregation graph leads to ViTs with significantly improved predictive performance; b) the graph measures of clustering  ...  to be much more advanced and optimized than the worm's neural networks.  ... 
arXiv:2206.11073v1 fatcat:hosyf6ze6zdwtdy5sdroostrqi

Jittor: a novel deep learning framework with meta-operators and unified graph execution

Shi-Min Hu, Dun Liang, Guo-Ye Yang, Guo-Wei Yang, Wen-Yang Zhou
2020 Science China Information Sciences  
Jittor: a novel deep learning framework with meta-operators and unified graph execution. Sci China Inf Sci, 2020, 63(12): 222103, https://doi.  ...  To manage metaoperators, Jittor uses a highly optimized way of executing computation graphs, which we call unified graph execution.  ...  We would like to thank anonymous reviewers for their helpful review comments, and Professor Ralph R. MARTIN for his useful suggestions and great help in paper writting.  ... 
doi:10.1007/s11432-020-3097-4 fatcat:t4ruztzhgbap5gvm4cliago5de

Graph Neural Networks for Wireless Communications: From Theory to Practice [article]

Yifei Shen, Jun Zhang, S.H. Song, Khaled B. Letaief
2022 arXiv   pre-print
For design guidelines, we propose a unified framework that is applicable to general design problems in wireless networks, which includes graph modeling, neural architecture design, and theory-guided performance  ...  For theoretical guarantees, we prove that GNNs achieve near-optimal performance in wireless networks with much fewer training samples than traditional neural architectures.  ...  The authors are with the Department of Electronic and Computer Engineering, Hong Kong University of Science and Technology, Hong Kong (E-mail:, {eejzhang, eeshsong eekhaled}  ... 
arXiv:2203.10800v1 fatcat:6o3jz7n3pjh3zhg76p3u3l2ugm

Sampling and Recovery of Graph Signals based on Graph Neural Networks [article]

Siheng Chen and Maosen Li and Ya Zhang
2020 arXiv   pre-print
We propose interpretable graph neural networks for sampling and recovery of graph signals, respectively.  ...  compared to previous neural-network-based sampling and recovery, the proposed methods are designed through exploiting specific graph properties and provide interpretability.  ...  With the attention vector a, we assign an trainable importance to each vertex, provide a path for backpropagation to flow gradients, and unify the training of a graph neural sampling module and a graph  ... 
arXiv:2011.01412v1 fatcat:vu4ozdtrxrfqxmzkyaffyj4wfu

UDRN: Unified Dimensional Reduction Neural Network for Feature Selection and Feature Projection [article]

Zelin Zang and Yongjie Xu and Yulan Geng and Siyuan Li and Stan Z. Li
2022 arXiv   pre-print
In this work, we develop a unified framework, Unified Dimensional Reduction Neural-network~(UDRN), that integrates FS and FP in a compatible, end-to-end way.  ...  FS and FP are traditionally incompatible categories; thus, they have not been unified into an amicable framework.  ...  Our contributions are summarized as follows: • Unified FS&FP problem and a novel neural network framework: We propose the problem of FS&FP with a unified objective and design a neural network framework  ... 
arXiv:2207.03809v1 fatcat:fd44fynvjne57kbvv3c6gahy6u

Chainer: A Deep Learning Framework for Accelerating the Research Cycle [article]

Seiya Tokui, Ryosuke Okuta, Takuya Akiba, Yusuke Niitani, Toru Ogawa, Shunta Saito, Shuji Suzuki, Kota Uenishi, Brian Vogel, Hiroyuki Yamazaki Vincent
2019 arXiv   pre-print
Software frameworks for neural networks play a key role in the development and application of deep learning methods.  ...  Chainer provides acceleration using Graphics Processing Units with a familiar NumPy-like API through CuPy, supports general and dynamic models in Python through Define-by-Run, and also provides add-on  ...  ACKNOWLEDGMENTS This work could not be achieved without the help of all the contributors and the feedback from users of Chainer, CuPy, ChainerCV, and related projects.  ... 
arXiv:1908.00213v1 fatcat:abyp464b7vagndtiwcet5cagtq
« Previous Showing results 1 — 15 out of 16,063 results