Global ETD Search
Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.
Results
Showing 1 to 20 of 75 for “"graph convolution"”.
-
Anomaly Detection from Videos using Graph Convolution Networks
… low-level features or leveraging the advent of convolutional neural networks or, more recently, image transformers. We propose a novel semi-supervised learning method to detect anomalies from a pedestrian dataset by representing each frame in the video feed as a graph. We create a graph …
-
Predicting crowd trajectories using deep Graph Convolution Neural Networks
… states. We propose the Social Spatio-Temporal Graph Convolutional Neural Network (Social-STGCNN), which substitutes the need of aggregation methods by modeling the interactions as a graph. Our results show an improvement over the state of art by 20% on the Final Displacement Error (FDE) and an …
-
Forecasting retweet count during elections using graph convolution neural networks
Indiana University-Purdue University Indianapolis (IUPUI)
-
Improving Deep Reinforcement Learning Using Graph Convolution and Visual Domain Transfer
… data representations (e.g. image based or graph based) 2) The reality gap issue in DRL. That is, how to effectively transfer a policy learned in a simulator to the real world. This thesis makes several novel contributions to the field of DRL which are outlined sequentially in the following. …
-
Detecting Public Transit Service Disruptions Using Social Media Mining and Graph Convolution
… and a couple different approaches that use Graph Neural Networks to identify transit disruption related information in Tweets from a live Twitter stream related to the Washington Metropolitan Area Transit Authority (WMATA) metro system. After developing three different models, a Dynamic …
-
DEEP LEARNING ON GRAPH-STRUCTURED DATA
Graph Neural Networks (GNNs) are generalized deep learning methods for processing graph-structured data. While GNNs have been widely utilized for various real-world tasks, such as social network analysis and drug discovery, their use in directed graphs is still in its infancy due to limitations in …
-
Baltymų molekulinių struktūrų sąveikų regionų nusakymas naudojant giliojo mokymosi metodus /
… study a deep learning method was developed using graph convolution networks to determine the binding sites in protein molecular structures. To achieve this, the graph deep learning algorithm was modelled using graph attention networks in PyTorch Geometric library. Data used for training, …
-
Resource-Efficient Machine Learning with Limited Data and Computation
… We focus on two machine learning models: graph neural networks and large language models. For graph neural networks, we identify their high cost stemming from graph data collection and graph neural network training. (i) To counteract scarce annotated graph datasets, we pro-posed an …
-
Machine Learning Driven Fine-Grained Spatial-Temporal Cellular User Traffic Prediction
… driven by machine learning algorithms. A deep graph convolution network, named GCGRN, is proposed. It is a novel combination of the graph convolution network (GCN) and gated recurrent units (GRU), which exploits graphical neural network to learn an efficient spatial-temporal model from a user's …
-
Machine Learning-Based Prediction of Sites of Metabolism in Drugs: Exploring Feature Extraction Methods on Molecular Graphs
… tasks. Seeing molecules as attributed graphs, where atoms correspond to nodes and bonds correspond to edges, the SOM prediction problem can be formulated as a node classification task. We compare two methods of extracting node features from molecular graphs: a standard fingerprint …
-
Spatial-Temporal Data Modeling with Graph Neural Networks
Spatial-temporal graph modeling is an important task to analyze the spatial relations and temporal trends of components in a system. It aims to model the dynamic node-level inputs by assuming inter-dependency between connected nodes. A basic assumption behind spatial-temporal graph modeling is that …
-
Deep learning architectures for 2D and 3D scene perception
… cloud information. We developed PointGrad, a new graph convolution gradient operator for capturing structural relationships, that encoded point-based directional gradients into a high-dimensional multiscale tensor space. Using the Point- Grad operator with graph convolution on scattered irregular …
-
Deep Learning Models for Traffic Prediction in Urban Transport Networks.
… a deep learning model (named SAGCN-SST) based on Graph Convolution Network (GCN) and Attention mechanism for multiple traffic speed prediction on large-scale road networks. This SAGCN- SST is able to capture dynamic-spatial and temporal features for the final prediction. The predicted traffic …
-
Protein interface prediction using graph convolutional networks
… a revolution of sorts with the emergence of convolutional neural networks as the leading method of choice for a wide swath of tasks. Enabled by large quantities of data and the increasing power and availability of computing resources, convolutional neural networks efficiently detect patterns …
-
On the Efficiency in Neural Networks: Exposing Unnecessary Computation
… on various tasks such as image recognition, graph classification, and reading comprehension. With tremendous computation, researchers and engineers are able to train unprecedentedly large models and achieve superhuman performance on several benchmarks. While the performance (or accuracy) of …
-
ALGORITHMIC IMPROVEMENTS FOR DIFFERENTIALLY PRIVATE LEARNING
… enhance accuracy. GCON enables edge-level DP for graph learning through objective perturbation, preserving graph convolution operations while providing tight sensitivity bounds. SaGD offers a rigorous node-level DP framework for graph learning, combining sensitivity-controlled message propagation …
-
Graph Neural Networks: A Feature and Structure Learning Approach
… in learning from more generic data such as graphs. In convolutional neural networks (CNNs), for example, the trainable local filters enable the automatic extraction of high-level features. The computation with filters requires a fixed number of ordered units in the receptive fields. However, …
-
Graph neural networks for mobile computer vision
… two research efforts focused on using vision graph neural networks (ViGs) as an alternative to convolutional neural networks (CNNs) and vision transformers (ViTs) for mobile vision tasks. We first propose MobileViG, a CNN-ViG-based hybrid architecture designed for mobile applications. To …
-
Geo-AI for wetland classification and evolutionary analysis
… this paper proposes the Unet-processed Mobile Graph Convolution network (UMGC) for analyzing hyperspectral images (HSI). UMGC employs Mobile Graph Convolution (MGC) to construct dynamic graphs more efficiently than traditional KNN-based methods. This approach is particularly suited for …
-
VRAG: Region Attention Graph for Content-Based Video Retrieval
… this paper, we introduce Video Region Attention Graph Networks (VRAG) that improves the state-of-the-art of video-level methods. We represent videos at a finer granularity viaregion-level features and encode video spatio-temporal dy-namics through region-level relations. Our VRAG capturesthe …
Page 1 of 4