Interpreting the Basis Path Set in Neural Networks

Zhu, Juanping; Meng, Qi; Chen, Wei; Ma, Zhiming

doi:10.1007/s11424-020-0112-y

Cited by 4 publications

(3 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As the healthcare domain is known to be critic and full of complicated scenarios that do not forgive mistakes, one accurate way to perform a deep learning technique is by preserving the model rationality [ 82 ]. Although model oriented [ 83 ] and example-based approaches [ 84 ] have shown an explainable independency level and an input dependent optimization respectively, they both position the problem of clarifying DNNs within a barrier of high interpretability but low accuracy, and vice versa.…”

Section: Methodsmentioning

confidence: 99%

An investigation into the deep learning approach in sentimental analysis using graph-based theories

Kentour

2021

PLoS ONE

View full text Add to dashboard Cite

Sentiment analysis is a branch of natural language analytics that aims to correlate what is expressed which comes normally within unstructured format with what is believed and learnt. Several attempts have tried to address this gap (i.e., Naive Bayes, RNN, LSTM, word embedding, etc.), even though the deep learning models achieved high performance, their generative process remains a “black-box” and not fully disclosed due to the high dimensional feature and the non-deterministic weights assignment. Meanwhile, graphs are becoming more popular when modeling complex systems while being traceable and understood. Here, we reveal that a good trade-off transparency and efficiency could be achieved with a Deep Neural Network by exploring the Credit Assignment Paths theory. To this end, we propose a novel algorithm which alleviates the features’ extraction mechanism and attributes an importance level of selected neurons by applying a deterministic edge/node embeddings with attention scores on the input unit and backward path respectively. We experiment on the Twitter Health News dataset were the model has been extended to approach different approximations (tweet/aspect and tweets’ source levels, frequency, polarity/subjectivity), it was also transparent and traceable. Moreover, results of comparing with four recent models on same data corpus for tweets analysis showed a rapid convergence with an overall accuracy of ≈83% and 94% of correctly identified true positive sentiments. Therefore, weights can be ideally assigned to specific active features by following the proposed method. As opposite to other compared works, the inferred features are conditioned through the users’ preferences (i.e., frequency degree) and via the activation’s derivatives (i.e., reject feature if not scored). Future direction will address the inductive aspect of graph embeddings to include dynamic graph structures and expand the model resiliency by considering other datasets like SemEval task7, covid-19 tweets, etc.

show abstract

Section: Methodsmentioning

confidence: 99%

An investigation into the deep learning approach in sentimental analysis using graph-based theories

Kentour

2021

PLoS ONE

View full text Add to dashboard Cite

show abstract

“…Informally, a path operation can be defined as an action that can be applied over a path or collection of paths. Examples of path operations are adding a path to a graph, delete a path from a graph and join two paths [106]. Current graph database systems are studying and implementing basic graph operations.…”

Section: P: Path Operationmentioning

confidence: 99%

Path Querying in Graph Databases: A Systematic Mapping Study

García,

Angles

2024

IEEE Access

View full text Add to dashboard Cite

Path querying refers to the evaluation of path queries in a graph database. New research in this topic is crucial for the development of graph database systems as path queries are associated with relevant use-cases and application domains. The aim of this article is to identify and establish what is currently known about path querying in graph databases. To achieve this, we conducted a systematic mapping study (SMS) in which we explored four digital libraries and collected research papers published from 1970 to 2022. These articles were filtered, classified and analyzed to extract quantitative and qualitative information which is presented in this article. Additionally, we provide a concise description of keywords, use-cases and application domains associated with path querying in graph databases.

show abstract

“…In these structured neural networks, a particularly important one called deep convolutional neural networks (DCNNs) has achieved state-of-the-art performance in many domains [2][3][4]. Normally, multichannel convolution is used, and the resulting multichannel deep convolutional neural networks (MDCNNs) have also achieved excellent performances in classification [5,6], natural language processing [7], biological [8][9][10], and many other domains [11][12][13].…”

Section: Introductionmentioning

confidence: 99%

Error Bounds for Approximations Using Multichannel Deep Convolutional Neural Networks with Downsampling

Liu

Hou

2023

Journal of Applied Mathematics

View full text Add to dashboard Cite

Deep learning with specific network topologies has been successfully applied in many fields. However, what is primarily called into question by people is its lack of theoretical foundation investigations, especially for structured neural networks. This paper theoretically studies the multichannel deep convolutional neural networks equipped with the downsampling operator, which is frequently used in applications. The results show that the proposed networks have outstanding approximation and generalization ability of functions from ridge class and Sobolev space. Not only does it answer an open and crucial question of why multichannel deep convolutional neural networks are universal in learning theory, but it also reveals the convergence rates.

show abstract

Interpreting the Basis Path Set in Neural Networks

Cited by 4 publications

References 15 publications

An investigation into the deep learning approach in sentimental analysis using graph-based theories

An investigation into the deep learning approach in sentimental analysis using graph-based theories

Path Querying in Graph Databases: A Systematic Mapping Study

Error Bounds for Approximations Using Multichannel Deep Convolutional Neural Networks with Downsampling

Contact Info

Product

Resources

About