Improved Explainability of Capsule Networks: Relevance Path by Agreement

Shahroudnejad, Atefeh; Mohammadi, Arash; Plataniotis, Konstantinos N.

doi:10.48550/arxiv.1802.10204

Cited by 3 publications

(4 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…CapsNet proposes the idea of equivalence instead of invariance and encapsulating pose information (such as scaling and rotation) and other instantiation parameters using capsules' activity vectors. Paper [69] analyzes CapsNet's nested architecture (Fig. 8) and verifies its explainable properties.…”

Section: B Decision Analysismentioning

confidence: 91%

“…The information post shows the basics of explainability concept through an illustrative example. The first figure illustrates the black-box concept, where the input is an image and the network prediction is a single word (e.g., face) [69]. As can be seen, such a single output provides no evidence for confirming the truth of predictions or rejecting incorrect predictions without having access to the ground-truth.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Survey on Understanding, Visualizations, and Explanation of Deep Neural Networks

Shahroudnejad¹

2021

Preprint

Self Cite

View full text Add to dashboard Cite

Recent advancements in machine learning and signal processing domains have resulted in an extensive surge of interest in Deep Neural Networks (DNNs) due to their unprecedented performance and high accuracy for different and challenging problems of significant engineering importance. However, when such deep learning architectures are utilized for making critical decisions such as the ones that involve human lives (e.g., in control systems and medical applications), it is of paramount importance to understand, trust, and in one word "explain" the argument behind deep models' decisions. In many applications, artificial neural networks (including DNNs) are considered as black-box systems, which do not provide sufficient clue on their internal processing actions. Although some recent efforts have been initiated to explain the behaviors and decisions of deep networks, explainable artificial intelligence (XAI) domain, which aims at reasoning about the behavior and decisions of DNNs, is still in its infancy. The aim of this paper is to provide a comprehensive overview on Understanding, Visualization, and Explanation of the internal and overall behavior of DNNs.

show abstract

Section: B Decision Analysismentioning

confidence: 91%

Section: Introductionmentioning

confidence: 99%

A Survey on Understanding, Visualizations, and Explanation of Deep Neural Networks

Shahroudnejad¹

2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Since the publication of the Dynamic Routing and the Cap-sNet, several works have emerged improving the algorithm or the architecture and experimenting with the power of CapsNet in other scenarios, applications, and datasets. Shahroudnejad, Mohammadi, and Plataniotis [6] presented an analysis of the explainability of CapsNet, showing that it has properties which help understand and explain its behavior. Jaiswal et al [7] used the CapsNet in a Generative Adversarial Network (GAN) and showed that it can achieve lower error rates than simple CNN.…”

Section: A Capsnetmentioning

confidence: 99%

The Multi-Lane Capsule Network (MLCN)

Rosario,

Borin,

Breternitz

2019

Preprint

View full text Add to dashboard Cite

We introduce Multi-Lane Capsule Networks (MLCN), which are a separable and resource efficient organization of Capsule Networks (CapsNet) [1] that allows parallel processing while achieving high accuracy at reduced cost. A MLCN is composed of a number of (distinct) parallel lanes, each contributing to a dimension of the result, trained using the routing-by-agreement organization of CapsNet. Our results indicate similar accuracy with a much-reduced cost in number of parameters for the Fashion-MNIST and Cifar10 datasets. They also indicate that the MLCN outperforms the original CapsNet when using a proposed novel configuration for the lanes. MLCN also has faster training and inference times, being more than twofold faster than the original CapsNet in the same accelerator.

show abstract

“…Since the initial publication, however, multiple improvements were proposed and the concept has been evolving. Shahroudnejad, Mohammadi, and Plataniotis [10] presented an analysis of the explainability of CapsNet, showing that it has properties to help understand and explain its behavior. Jaiswal et al [11] used the CapsNet in a Generative Adversarial Network (GAN) and showed that it can achieve lower error rates than the simple CNN.…”

Section: A Capsule Networkmentioning

confidence: 99%

Efficiency and Scalability of Multi-Lane Capsule Networks (MLCN)

Rosario¹,

Breternitz²,

Borin³

2019

Preprint

View full text Add to dashboard Cite

Some Deep Neural Networks (DNN) have what we call lanes, or they can be reorganized as such. Lanes are paths in the network which are data-independent and typically learn different features or add resilience to the network. Given their data-independence, lanes are amenable for parallel processing. The Multi-lane CapsNet (MLCN) is a proposed reorganization of the Capsule Network which is shown to achieve better accuracy while bringing highly-parallel lanes. However, the efficiency and scalability of MLCN had not been systematically examined.In this work, we study the MLCN network with multiple GPUs finding that it is 2x more efficient than the original CapsNet when using model-parallelism. Further, we present the load balancing problem of distributing heterogeneous lanes in homogeneous or heterogeneous accelerators and show that a simple greedy heuristic can be almost 50% faster than a naïve random approach.Index Terms-deep learning capsule network multi-lane I. INTRODUCTIONSeveral approaches to the distributed model parallelization of Deep Neural Networks (DNN) have concentrated in their depth [1]-[3], but DNNs can also be organized in a way to be parallelized in their width [4]. The DNN architecture may be organized into distinct neural network lanes [5]. This creates separable and resource efficient data-independent paths in the network that can be used to learn different features or add resilience to the network. Examples of neural networks with lanes are the Google Inception [6], [7] and the Multi-lane Capsule Network (MLCN) [5]. As these lanes are data-independent they can be (1) processed in parallel and (2) specialized for distinct computational targets (CPUs, GPU, FPGAs, and cloud), as well as resourceconstrained mobile and IoT targets, leading to opportunities and challenges. Recently, our research focus was on Multi-Lane Capsule Networks (MLCN), which are a separable and resource efficient organization of Capsule Networks (CapsNet) This work was supported in part by CAPES/Brasil (Finance Code 001), by CNPq (313012/2017-2), and by Fapesp (CCES 2013/08293-7). We would like to thank Google Cloud Platform for a grant to run our experiments.V. M. do Rosario, is a Ph.D.

show abstract

Improved Explainability of Capsule Networks: Relevance Path by Agreement

Cited by 3 publications

References 0 publications

A Survey on Understanding, Visualizations, and Explanation of Deep Neural Networks

A Survey on Understanding, Visualizations, and Explanation of Deep Neural Networks

The Multi-Lane Capsule Network (MLCN)

Efficiency and Scalability of Multi-Lane Capsule Networks (MLCN)

Contact Info

Product

Resources

About