Re-training Deep Neural Networks to Facilitate Boolean Concept Extraction

González, Camila; Mencía, Eneldo Loza; Fürnkranz, Johannes

doi:10.1007/978-3-319-67786-6_10

Cited by 8 publications

(7 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…More recently, Zilke et al (2016) proposed an algorithm that extracts decision trees per layer which can be merged into one rule set for the complete NN. González et al (2017) improve on this algorithm by polarizing real-valued activations and pruning weights through retraining. Both rely on the C4.5 ( Quinlan, 2014 ) decision tree algorithm for rule extraction.…”

Section: Related Workmentioning

confidence: 99%

Rule Extraction From Binary Neural Networks With Convolutional Rules for Model Validation

Burkhardt

Brugger

Wagner

et al. 2021

Front. Artif. Intell.

View full text Add to dashboard Cite

Classification approaches that allow to extract logical rules such as decision trees are often considered to be more interpretable than neural networks. Also, logical rules are comparatively easy to verify with any possible input. This is an important part in systems that aim to ensure correct operation of a given model. However, for high-dimensional input data such as images, the individual symbols, i.e. pixels, are not easily interpretable. Therefore, rule-based approaches are not typically used for this kind of high-dimensional data. We introduce the concept of first-order convolutional rules, which are logical rules that can be extracted using a convolutional neural network (CNN), and whose complexity depends on the size of the convolutional filter and not on the dimensionality of the input. Our approach is based on rule extraction from binary neural networks with stochastic local search. We show how to extract rules that are not necessarily short, but characteristic of the input, and easy to visualize. Our experiments show that the proposed approach is able to model the functionality of the neural network while at the same time producing interpretable logical rules. Thus, we demonstrate the potential of rule-based approaches for images which allows to combine advantages of neural networks and rule learning.

show abstract

Section: Related Workmentioning

confidence: 99%

Rule Extraction From Binary Neural Networks With Convolutional Rules for Model Validation

Burkhardt

Brugger

Wagner

et al. 2021

Front. Artif. Intell.

View full text Add to dashboard Cite

show abstract

“…From Equation (8) and the condition of Equation (2), Lagrange's undetermined multiplier α is determined, and Equation (8) is rewritten as Equation (5). Table 1 shows the detailed settings for the experiments described in sections 3.1, 3.2, and 3.3.…”

Section: Appendix a Proof Of Em Algorithm For Community Detectionmentioning

confidence: 99%

“…• Approach B: Training interpretable layered neural networks -This approach is used to devise the training methods so that the trained layered neural network is represented by an interpretable function ( [7,8]). For instance, training a network has been proposed as a way of performing an affine transformation or one that can be represented in a rule-based manner.…”

Section: Introductionmentioning

confidence: 99%

Understanding community structure in layered neural networks

Watanabe¹,

Hiramatsu²,

Kashino³

2019

Neurocomputing

View full text Add to dashboard Cite

A layered neural network is now one of the most common choices for the prediction or recognition of high-dimensional practical data sets, where the relationship between input and output data is complex and cannot be represented well by simple conventional models. Its * Email address: watanabe.chihiro@lab.ntt.co.jp our previous methods, by defining the effect of each input dimension on a community, and the effect of a community on each output dimension. We show experimentally that our proposed method can reveal the role of each part of a layered neural network by applying the neural networks to three types of data sets, extracting communities from the trained network, and applying the proposed method to the community structure.The methods described in the above related studies have enabled us to acquire knowledge about the mechanism of a whole layered neural network or that of each unit in a layered neural network, however, they cannot capture the global network structure of a trained layered neural network.Therefore, we proposed the first method for extracting a global layered neural network structure by applying network analysis to a trained layered neural network ([27, 26, 28, 25]). By using these methods, we can decompose units in a layered neural network into groups or communities, where all units have similar connection patterns with adjacent layers. The extracted community structure of a layered neural network provides information about the roles of each part of a layered neural network.

show abstract

“…Consequently, the interest in methods that make learned models more interpretable has increased with the success of deep learning. Some research has been devoted to trying to convert such arcane models to more interpretable rule-based (Andrews et al, 1995) or tree-based models (Frosst and Hinton, 2017), which may be facilitated with appropriate neural network training techniques (González et al, 2017). Instead of making the entire model interpretable, methods like LIME are able to provide local explanations for inscrutable models, allowing a trade-off between fidelity to the original model with interpretability and complexity of the local model.…”

Section: Neural Network and Deep Learningmentioning

confidence: 99%

Advances in Machine Learning for the Behavioral Sciences

Kliegr

Bahník

Fürnkranz

2019

American Behavioral Scientist

Self Cite

View full text Add to dashboard Cite

The areas of machine learning and knowledge discovery in databases have considerably matured in recent years. In this article, we briefly review recent developments as well as classical algorithms that stood the test of time. Our goal is to provide a general introduction into different tasks such as learning from tabular data, behavioral data, or textual data, with a particular focus on actual and potential applications in behavioral sciences. The supplemental appendix to the article also provides practical guidance for using the methods by pointing the reader to proven software implementations. The focus is on R, but we also cover some libraries in other programming languages as well as systems with easy-to-use graphical interfaces.

show abstract

Re-training Deep Neural Networks to Facilitate Boolean Concept Extraction

Cited by 8 publications

References 17 publications

Rule Extraction From Binary Neural Networks With Convolutional Rules for Model Validation

Rule Extraction From Binary Neural Networks With Convolutional Rules for Model Validation

Understanding community structure in layered neural networks

Advances in Machine Learning for the Behavioral Sciences

Contact Info

Product

Resources

About