93,185 Hits in 3.8 sec

Learning to Route with Sparse Trajectory Sets---Extended Version [article]

Chenjuan Guo, Bin Yang, Jilin Hu, Christian S. Jensen
2018 arXiv   pre-print
This is an extended version of "Learning to Route with Sparse Trajectory Sets" [1], to appear in IEEE ICDE 2018.  ...  In the second step, we exploit the above graph-like structure to achieve a comprehensive trajectory-based routing solution.  ...  Transferring preferences among similar region edges: We adopt the idea of graph-based transduction learning [39] , [40] to transfer routing preferences from T-edges to similar Bedges.  ... 
arXiv:1802.07980v1 fatcat:7cxpegdhvrdwxlkqpp7zq4a2zq

DCG-Net: Dynamic Capsule Graph Convolutional Network for Point Clouds

Dena Bazazian, Dhananjay Nahata
2020 IEEE Access  
However, the key idea of this work is to apply a routing mechanism to find the similarities between all the points of a point cloud in order to build a graph by neighborhood aggregation from a raw data  ...  In DCG-Net, we applied a dynamic routing mechanism between capsules to find similar points based on the features from three different spaces to construct the graphs at the first layer, and then we updated  ... 
doi:10.1109/access.2020.3031812 fatcat:nhhj2yr4pzgbnesrslp5c4r2eu

From Cognitive Maps to Cognitive Graphs

Elizabeth R. Chrastil, William H. Warren, Robert Sutherland
2014 PLoS ONE  
In contrast to route knowledge, we find that the most frequent routes and detours to target locations had not been traveled during learning.  ...  We present evidence that this structure is similar to a labeled graph: a network of topological connections between places, labeled with local metric information.  ...  In contrast to route knowledge, we found that the most frequent routes and detours to target locations in the test phase had not been traveled during learning.  ... 
doi:10.1371/journal.pone.0112544 pmid:25389769 pmcid:PMC4229194 fatcat:c6idqxksvbckzn6o7wcxpwuabe

Different clones for different contexts: Hippocampal cognitive maps as higher-order graphs of a cloned HMM [article]

Nishad Gothoskar, J. Swaroop Guntupalli, Rajeev V Rikhye, Miguel Lázaro-Gredilla, Dileep George
2019 bioRxiv   pre-print
In our experiments, we use the cloned HMM for learning spatial and abstract representations.  ...  We show that inference and planning in the learned CHMM encapsulates many of the key properties of hippocampal cells observed in rodents and humans.  ...  After learning, the SR of CHMM transition matrix ( Fig 2C) showed representational similarity with the original graph.  ... 
doi:10.1101/745950 fatcat:v7jn5rvuozdhfgo5yhajdahn2u

Graph Routing between Capsules [article]

Yang Li, Wei Zhao, Erik Cambria, Suhang Wang, Steffen Eger
2021 arXiv   pre-print
Therefore, in this paper, we introduce a new capsule network with graph routing to learn both relationships, where capsules in each layer are treated as the nodes of a graph.  ...  Routing methods in capsule networks often learn a hierarchical relationship for capsules in successive layers, but the intra-relation between capsules in the same layer is less studied, while this intra-relation  ...  Conclusion & Future Works It is the first time to treat the capsule as a node in a graph, and our graph routing applies GCN to explore the relationship between capsules in the same layer.  ... 
arXiv:2106.11531v1 fatcat:cahjd7vyg5bgtf4xkhhyta6o4u

Generalizable Cross-Graph Embedding for GNN-based Congestion Prediction [article]

Amur Ghose, Vincent Zhang, Yingxue Zhang, Dong Li, Wulong Liu, Mark Coates
2021 arXiv   pre-print
In our framework, we introduce a superior alternative to obtain node embeddings that can generalize across netlist graphs using matrix factorization methods.  ...  By combining the learned embedding on top of the netlist with the GNNs, our method improves prediction performance, generalizes to new circuit lines, and is efficient in training, potentially saving over  ...  We begin by discussing similarity notions in the context of embedding learning on graphs. B.  ... 
arXiv:2111.05941v1 fatcat:va47o5r5z5hljgysbfwf5fllby

From Route Instructions to Landmark Graphs [article]

Christopher M Cervantes
2020 arXiv   pre-print
We propose the landmark graph generation task (creating landmark-based spatial representations from natural language) and introduce a fully end-to-end neural approach to generate these graphs.  ...  We evaluate our models on the SAIL route instruction dataset, as well as on a small set of real-world delivery instructions that we collected, and we show that our approach yields high quality results  ...  approach predicts landmark graphs with a high degree of similarity to the ground truth.  ... 
arXiv:2002.02012v1 fatcat:sikhvtt32nd6dgl4qttmo2bxky

GDDR: GNN-based Data-Driven Routing [article]

Oliver Hope, Eiko Yoneki
2021 arXiv   pre-print
We explore the feasibility of combining Graph Neural Network-based policy architectures with Deep Reinforcement Learning as an approach to problems in systems.  ...  As a case study, we take the idea of data-driven routing in intradomain traffic engineering, whereby the routing of data in a network can be managed taking into account the data itself.  ...  ACKNOWLEDGMENT The authors would like to thank Kai Fricke for his input on GDDR project. This research was partly funded by the Alan Turing Institute.  ... 
arXiv:2104.09919v1 fatcat:jac6pzq65bhsxfczix3xkub7um

Active and passive spatial learning in human navigation: Acquisition of graph knowledge

Elizabeth R. Chrastil, William H. Warren
2015 Journal of Experimental Psychology. Learning, Memory and Cognition  
Route and graph knowledge were assessed by walking in the maze corridors from a starting object to the remembered location of a test object, with frequent detours.  ...  In this study, we test the contributions of 3 components to topological graph and route knowledge: visual information, idiothetic information, and cognitive decision making.  ...  First, we predict that decision making will play a significant role in route and graph learning, in contrast to survey learning.  ... 
doi:10.1037/xlm0000082 pmid:25419818 fatcat:6zrvyz32avbyfmtuggrin52h7m

DemiNet: Dependency-Aware Multi-Interest Network with Self-Supervised Graph Learning for Click-Through Rate Prediction [article]

Yule Wang, Qiang Luo, Yue Ding, Dong Wang, Hongbo Deng
2021 arXiv   pre-print
To filter out noisy inter-item correlations and enhance the robustness of extracted interests, self-supervised interest learning is introduced to the above two steps.  ...  In this paper, we propose a novel model named DemiNet (short for DEpendency-Aware Multi-Interest Network}) to address the above two issues.  ...  To aggregate the multiple interests, route-specific interest experts and Confi-Net are introduced. We conduct extensive experiments on three real-world datasets.  ... 
arXiv:2109.12512v1 fatcat:vznmni5xfzhsnjkfnxh5kfyhsa

Humpty Dumpty: Putting iBGP Back Together Again [chapter]

Ashley Flavel, Jeremy McMahon, Aman Shaikh, Matthew Roughan, Nigel Bean
2009 Lecture Notes in Computer Science  
This technique has been successfully applied in a 'what-if' scenario and has future applications in the real-time analysis of routing decisions.  ...  picture of a network's routing state.  ...  Similar techniques can be applied to any iBGP topology if the topology can be abstracted to a reliance graph.  ... 
doi:10.1007/978-3-642-01399-7_5 fatcat:rjwpmyqekjccbfhn7j64vjcp3e

Helping the ineloquent farmers: Finding experts for questions with limited text in agricultural Q&A Communities

Xiaoxue Shen, Adele Lu Jia, Siqi Shen, Yong Dou
2020 IEEE Access  
INDEX TERMS Question and answering, question routing, network representation learning.  ...  We propose a novel approach based on graph neural network to accurately recommend for each question the users that are highly likely to answer it.  ...  Recently, many researchers have used graph neural networks to learn node representations, which aims to apply the deep neural network into the graph-structured data.  ... 
doi:10.1109/access.2020.2984342 fatcat:z3h3nkswyzeudid6r77n23nlx4

Learning to Delegate for Large-scale Vehicle Routing [article]

Sirui Li, Zhongxia Yan, Cathy Wu
2021 arXiv   pre-print
While previous heuristic or learning-based works achieve decent solutions on small problem instances of up to 100 cities, their performance deteriorates in large problems.  ...  This article presents a novel learning-augmented local search framework to solve large-scale VRP.  ...  FC network to obtain the input route-node features for the next graph attention layer, as each subproblem S corresponds to a single route r by construction as described in Subsection 4.1.  ... 
arXiv:2107.04139v2 fatcat:qsgdno7n5fgcza77hyueeslnp4

Packet Routing with Graph Attention Multi-agent Reinforcement Learning [article]

Xuan Mai, Quanzhi Fu, Yi Chen
2021 arXiv   pre-print
Considering the graph nature of the network topology, we design a multi-agent RL framework in combination with Graph Neural Network (GNN), tailored to the routing problem.  ...  In this paper, we develop a model-free and data-driven routing strategy by leveraging reinforcement learning (RL), where routers interact with the network and learn from the experience to make some good  ...  Moreover, all agents concurrently learn similar policies but not necessarily identical policies. V. CONCLUSION In this paper, we study the packet routing problem for communication networks.  ... 
arXiv:2107.13181v1 fatcat:zc2whz7cbbhdrd44nvo3zjxdfe

Non-Euclidean navigation

William H. Warren
2019 Journal of Experimental Biology  
The results are contrary to the Euclidean map hypothesis and support the cognitive graph hypothesis.  ...  What distinguishes such a cognitive graph from a metric cognitive map is that this local information is not embedded in a global coordinate system, so spatial knowledge is often geometrically inconsistent  ...  article, and to Bob Shaw, from whom I learned to find the right geometry for your problem.  ... 
doi:10.1242/jeb.187971 pmid:30728233 fatcat:id5xrrchencwthtpbzoo55fvt4
« Previous Showing results 1 — 15 out of 93,185 results