Nonlinear distance function learning using neural network: an iterative framework

2014 ◽  
Vol 74 (3) ◽  
pp. 671-688 ◽  
Author(s):  
Junying Chen ◽  
Haoyu Zeng ◽  
Na Fan
Keyword(s):  
Neural Network ◽  
Distance Function ◽  
Function Learning ◽  
Iterative Framework

Research on Dissolve Oxygen Modeling Based on Neural Network

2011 ◽  
Vol 287-290 ◽  
pp. 2640-2643
Author(s):  
Guo Dong Gao ◽  
Wen Xiao Zhang ◽  
Gong Zhi Yu ◽  
Jiang Hua Sui
Keyword(s):  
Neural Network ◽  
Dissolved Oxygen ◽  
Network Model ◽  
Neural Network Model ◽  
Bp Neural Network ◽  
Structure Learning ◽  
Learning Algorithm ◽  
Impact Factors ◽  
Function Learning ◽  
Bp Neural Network Model

The structure, characteristics and principles of BP neural network model are described in this paper. First, three impact factors of the dissolved oxygen are selected as the sample input of network, and then the parameters of BP neural network are selected, such as network structure, learning algorithm, output layer transfer function, learning rate and so on. Finally, the BP neural network model is established and trained, in order to approach compensate the effects of improves non-linearity. The simulation results show that BP neural network is practical and dependable in the field of dissolved oxygen modeling and has nice applied prospect.

scholarly journals Ontology optimization tactics via distance calculating

2016 ◽  
Vol 1 (1) ◽  
pp. 159-174 ◽  
Author(s):  
Yun Gao ◽  
Mohammad Reza Farahani ◽  
Wei Gao
Keyword(s):  
Similarity Measure ◽  
Distance Function ◽  
High Efficiency ◽  
Learning Algorithm ◽  
Ontology Mapping ◽  
Ontology Learning ◽  
Real Numbers ◽  
Function Learning ◽  
Engineering Applications ◽  
Learning Techniques

AbstractIn this article, we propose an ontology learning algorithm for ontology similarity measure and ontology mapping in view of distance function learning techniques. Using the distance computation formulation, all the pairs of ontology vertices are mapped into real numbers which express the distance of their corresponding vectors. The more distance between two vertices, the smaller similarity between their corresponding concepts. The stabilities of our learning algorithm are defined and several bounds are yielded via stability assumptions. The simulation experimental conclusions show that the new proposed ontology algorithm has high efficiency and accuracy in ontology similarity measure and ontology mapping in certain engineering applications.

scholarly journals An Iterative Scheme of Safe Reinforcement Learning for Nonlinear Systems via Barrier Certificate Generation

2021 ◽  
pp. 467-490
Author(s):  
Zhengfeng Yang ◽  
Yidan Zhang ◽  
Wang Lin ◽  
Xia Zeng ◽  
Xiaochao Tang ◽  
...  
Keyword(s):  
Neural Network ◽  
Reinforcement Learning ◽  
Nonlinear Systems ◽  
Deep Neural Network ◽  
Iterative Scheme ◽  
Synthesis Methods ◽  
Learning Approach ◽  
Bilinear Matrix Inequalities ◽  
Safety Constraints ◽  
Iterative Framework

AbstractIn this paper, we propose a safe reinforcement learning approach to synthesize deep neural network (DNN) controllers for nonlinear systems subject to safety constraints. The proposed approach employs an iterative scheme where a learner and a verifier interact to synthesize safe DNN controllers. The learner trains a DNN controller via deep reinforcement learning, and the verifier certifies the learned controller through computing a maximal safe initial region and its corresponding barrier certificate, based on polynomial abstraction and bilinear matrix inequalities solving. Compared with the existing verification-in-the-loop synthesis methods, our iterative framework is a sequential synthesis scheme of controllers and barrier certificates, which can learn safe controllers with adaptive barrier certificates rather than user-defined ones. We implement the tool SRLBC and evaluate its performance over a set of benchmark examples. The experimental results demonstrate that our approach efficiently synthesizes safe DNN controllers even for a nonlinear system with dimension up to 12.

Sign in / Sign up

Export Citation Format

Share Document