Dark Reciprocal-Rank: Teacher-to-student Knowledge Transfer from Self-localization Model to Graph-convolutional Neural Network

2021 ◽  
Author(s):  
Takeda Koji ◽  
Tanaka Kanji
Keyword(s):  
Neural Network ◽  
Knowledge Transfer ◽  
Convolutional Neural Network ◽  
Student Knowledge ◽  
Localization Model ◽  
Reciprocal Rank

scholarly journals Knowledge Transfer from Map to DNN: Use of Graph Convolutional Neural Network for Augmenting Visual Robot Self-localization System

2020 ◽  
Author(s):  
koji takeda ◽  
kanji tanaka
Keyword(s):  
Neural Network ◽  
Knowledge Transfer ◽  
Convolutional Neural Network ◽  
Large Scale ◽  
Graph Matching ◽  
Training Data ◽  
Localization System ◽  
Scene Graph ◽  
System P ◽  
Localization Model

Graph-based scene model has been receiving increasing attention as a flexible and descriptive scene model for visual robot self-localization. In a typical self-localization application, objects, object features, and object relationship in an environment map are described respectively by nodes, node features, and edges in a scene graph, which are then matched against a query scene graph by a graph matching engine. However, its overhead for computation, storage, and communication, is proportional to the number and feature dimensionality of graph nodes, and can be significant in large-scale applications. In this study, we observe that graph-convolutional neural network (GCN) has a potential to become an efficient tool to train and predict with a graph matching engine. However, it is non-trivial to translate a given visual feature to a proper graph feature that contributes to good self-localization performance. To address this issue, we introduce a new knowledge transfer (KT) framework, which introduces an arbitrary self-localization model as a teacher to train the student, GCN-based self-localization system. Our KT framework enables lightweight storage/communication by using compact teacher's output signals as training data. Results on RobotCar datasets show that the proposed method outperforms existing comparing methods as well as the teacher self-localization system.

scholarly journals Multitask Prediction of Site Selectivity in Aromatic C-H Functionalization Reactions

2019 ◽  
Author(s):  
Thomas J. Struble ◽  
Connor W. Coley ◽  
Klavs F. Jensen
Keyword(s):  
Neural Network ◽  
Drug Discovery ◽  
Convolutional Neural Network ◽  
High Selectivity ◽  
Accurate Prediction ◽  
Process Chemistry ◽  
Site Selectivity ◽  
Synthetic Routes ◽  
Validation Testing ◽  
Reciprocal Rank

Aromatic C-H functionalization reactions are an important part of the synthetic chemistry toolbox. Accurate prediction of site selectivity can be crucial for prioritizing target compounds and synthetic routes in both drug discovery and process chemistry. However, selectivity may be highly dependent on subtle electronic and steric features of the substrate. We report a generalizable approach to prediction of site selectivity that is accomplished using a graph-convolutional neural network for the multitask prediction of 123 C-H functionalization tasks. In an 80/10/10 training/validation/testing pseudo-time split of about 58,000 aromatic C-H functionalization reactions from the Reaxys database, the model achieves a mean reciprocal rank of 92%. Once trained, inference requires approximately 200 ms per compound to provide quantitative likelihood scores for each task. This approach and model allow a chemist to quickly determine which C-H functionalization reactions-if any-might proceed with high selectivity.

scholarly journals Multitask Prediction of Site Selectivity in Aromatic C-H Functionalization Reactions

2019 ◽  
Author(s):  
Thomas J. Struble ◽  
Connor W. Coley ◽  
Klavs F. Jensen
Keyword(s):  
Neural Network ◽  
Drug Discovery ◽  
Convolutional Neural Network ◽  
High Selectivity ◽  
Accurate Prediction ◽  
Process Chemistry ◽  
Site Selectivity ◽  
Synthetic Routes ◽  
Validation Testing ◽  
Reciprocal Rank

Aromatic C-H functionalization reactions are an important part of the synthetic chemistry toolbox. Accurate prediction of site selectivity can be crucial for prioritizing target compounds and synthetic routes in both drug discovery and process chemistry. However, selectivity may be highly dependent on subtle electronic and steric features of the substrate. We report a generalizable approach to prediction of site selectivity that is accomplished using a graph-convolutional neural network for the multitask prediction of 123 C-H functionalization tasks. In an 80/10/10 training/validation/testing pseudo-time split of about 58,000 aromatic C-H functionalization reactions from the Reaxys database, the model achieves a mean reciprocal rank of 92%. Once trained, inference requires approximately 200 ms per compound to provide quantitative likelihood scores for each task. This approach and model allow a chemist to quickly determine which C-H functionalization reactions-if any-might proceed with high selectivity.

CONVOLUTIONAL NEURAL NETWORK BASED ALGORITHM FOR CECUM ACHIEVEMENT CONFIRMATION

2020 ◽  
Author(s):  
S Kashin ◽  
D Zavyalov ◽  
A Rusakov ◽  
V Khryashchev ◽  
A Lebedev
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network

Non-Blind Image Deconvolution Based on “Ringing” Removal Using Convolutional Neural Network

2020 ◽  
Vol 2020 (10) ◽  
pp. 181-1-181-7
Author(s):  
Takahiro Kudo ◽  
Takanori Fujisawa ◽  
Takuro Yamaguchi ◽  
Masaaki Ikehara
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Network Architecture ◽  
Large Scale ◽  
Blind Deconvolution ◽  
Training Dataset ◽  
Image Deconvolution ◽  
Classic Problem ◽  
Key Points ◽  
Blind Image

Image deconvolution has been an important issue recently. It has two kinds of approaches: non-blind and blind. Non-blind deconvolution is a classic problem of image deblurring, which assumes that the PSF is known and does not change universally in space. Recently, Convolutional Neural Network (CNN) has been used for non-blind deconvolution. Though CNNs can deal with complex changes for unknown images, some CNN-based conventional methods can only handle small PSFs and does not consider the use of large PSFs in the real world. In this paper we propose a non-blind deconvolution framework based on a CNN that can remove large scale ringing in a deblurred image. Our method has three key points. The first is that our network architecture is able to preserve both large and small features in the image. The second is that the training dataset is created to preserve the details. The third is that we extend the images to minimize the effects of large ringing on the image borders. In our experiments, we used three kinds of large PSFs and were able to observe high-precision results from our method both quantitatively and qualitatively.

Sign in / Sign up

Export Citation Format

Share Document