Visualizing the Hidden Activity of Artificial Neural Networks

Publikation: Beiträge in ZeitschriftenZeitschriftenaufsätzeForschungbegutachtet

Standard

Visualizing the Hidden Activity of Artificial Neural Networks. / Rauber, Paulo E.; Fadel, Samuel G.; Falcão, Alexandre X. et al.
in: IEEE Transactions on Visualization and Computer Graphics, Jahrgang 23, Nr. 1, 7539329, 01.2017, S. 101-110.

Publikation: Beiträge in ZeitschriftenZeitschriftenaufsätzeForschungbegutachtet

Harvard

APA

Vancouver

Rauber PE, Fadel SG, Falcão AX, Telea AC. Visualizing the Hidden Activity of Artificial Neural Networks. IEEE Transactions on Visualization and Computer Graphics. 2017 Jan;23(1):101-110. 7539329. doi: 10.1109/TVCG.2016.2598838

Bibtex

@article{65c0ec990e9241a0b69a8ffd4695998f,
title = "Visualizing the Hidden Activity of Artificial Neural Networks",
abstract = "In machine learning, pattern classification assigns high-dimensional vectors (observations) to classes based on generalization from examples. Artificial neural networks currently achieve state-of-the-art results in this task. Although such networks are typically used as black-boxes, they are also widely believed to learn (high-dimensional) higher-level representations of the original observations. In this paper, we propose using dimensionality reduction for two tasks: visualizing the relationships between learned representations of observations, and visualizing the relationships between artificial neurons. Through experiments conducted in three traditional image classification benchmark datasets, we show how visualization can provide highly valuable feedback for network designers. For instance, our discoveries in one of these datasets (SVHN) include the presence of interpretable clusters of learned representations, and the partitioning of artificial neurons into groups with apparently related discriminative roles.",
keywords = "algorithm understanding, Artificial neural networks, dimensionality reduction, Informatics, Business informatics",
author = "Rauber, {Paulo E.} and Fadel, {Samuel G.} and Falc{\~a}o, {Alexandre X.} and Telea, {Alexandru C.}",
year = "2017",
month = jan,
doi = "10.1109/TVCG.2016.2598838",
language = "English",
volume = "23",
pages = "101--110",
journal = "IEEE Transactions on Visualization and Computer Graphics",
issn = "1077-2626",
publisher = "IEEE - Institute of Electrical and Electronics Engineers Inc.",
number = "1",

}

RIS

TY - JOUR

T1 - Visualizing the Hidden Activity of Artificial Neural Networks

AU - Rauber, Paulo E.

AU - Fadel, Samuel G.

AU - Falcão, Alexandre X.

AU - Telea, Alexandru C.

PY - 2017/1

Y1 - 2017/1

N2 - In machine learning, pattern classification assigns high-dimensional vectors (observations) to classes based on generalization from examples. Artificial neural networks currently achieve state-of-the-art results in this task. Although such networks are typically used as black-boxes, they are also widely believed to learn (high-dimensional) higher-level representations of the original observations. In this paper, we propose using dimensionality reduction for two tasks: visualizing the relationships between learned representations of observations, and visualizing the relationships between artificial neurons. Through experiments conducted in three traditional image classification benchmark datasets, we show how visualization can provide highly valuable feedback for network designers. For instance, our discoveries in one of these datasets (SVHN) include the presence of interpretable clusters of learned representations, and the partitioning of artificial neurons into groups with apparently related discriminative roles.

AB - In machine learning, pattern classification assigns high-dimensional vectors (observations) to classes based on generalization from examples. Artificial neural networks currently achieve state-of-the-art results in this task. Although such networks are typically used as black-boxes, they are also widely believed to learn (high-dimensional) higher-level representations of the original observations. In this paper, we propose using dimensionality reduction for two tasks: visualizing the relationships between learned representations of observations, and visualizing the relationships between artificial neurons. Through experiments conducted in three traditional image classification benchmark datasets, we show how visualization can provide highly valuable feedback for network designers. For instance, our discoveries in one of these datasets (SVHN) include the presence of interpretable clusters of learned representations, and the partitioning of artificial neurons into groups with apparently related discriminative roles.

KW - algorithm understanding

KW - Artificial neural networks

KW - dimensionality reduction

KW - Informatics

KW - Business informatics

UR - http://www.scopus.com/inward/record.url?scp=84998953869&partnerID=8YFLogxK

U2 - 10.1109/TVCG.2016.2598838

DO - 10.1109/TVCG.2016.2598838

M3 - Journal articles

C2 - 27875137

AN - SCOPUS:84998953869

VL - 23

SP - 101

EP - 110

JO - IEEE Transactions on Visualization and Computer Graphics

JF - IEEE Transactions on Visualization and Computer Graphics

SN - 1077-2626

IS - 1

M1 - 7539329

ER -

DOI

Zuletzt angesehen

Publikationen

  1. Perfectly nested or significantly nested - an important difference for conservation management
  2. Combined MRI-PET dissects dynamic changes in plant structures and functions
  3. Advisory systems in pluralistic knowledge societies:
  4. A Voxel-based technique to estimate the volume of trees from terrestrial laser scanner data
  5. A longitudinal multilevel CFA-MTMM model for interchangeable and structurally different methods
  6. Exploiting ConvNet diversity for flooding identification
  7. How Much Home Office is Ideal? A Multi-Perspective Algorithm
  8. Phosphorus uptake from struvite is modulated by the nitrogen form applied
  9. How can problems be turned into something good? The role of entrepreneurial learning and error mastery orientation
  10. Scaling-based Least Squares Methods with Implemented Kalman filter Approach for Nano-Parameters Identification
  11. Understanding Low-Code Evolution, Adoption and Ecosystem for Software Development
  12. Reducing mean tardiness in a flexible job shop containing AGVs with optimized combinations of sequencing and routing rules
  13. Artificial intelligence in songwriting and composing - perspectives and challenges in creative practices
  14. Errors in Training Computer Skills
  15. Learning and Re-learning from net- based cooperative learning discourses
  16. Combining flatness based feedforward action with a fractional PI regulator to control the intake valve engine
  17. Offline question answering over linked data using limited resources
  18. Learning in the "Third Space"
  19. Schooling, local knowledge and working memory
  20. Unraveling Privacy Concerns in Complex Data Ecosystems with Architectural Thinking
  21. Using Daily Stretching to Counteract Performance Decreases as a Result of Reduced Physical Activity—A Controlled Trial
  22. Fusion of knowledge bases for better navigation of wheeled mobile robotic group with 3D TVS
  23. Reciprocal Relationships Between Dispositional Optimism and Work Experiences
  24. Teachers’ temporary support and worked-out examples as elements of scaffolding in mathematical modeling
  25. Introduction
  26. Application of feedforward artificial neural network in Muskingum flood routing
  27. Who can nudge for sustainable development? How nudge source renders dynamic norms (in-)effective in eliciting sustainable behavior
  28. Toward a methodical framework for comprehensively assessing forest multifunctionality
  29. Comparative study on the dehydrogenation properties of TiCl4-doped LiAlH4 using different doping techniques
  30. The Low-Code Phenomenon: Mapping the Intellectual Structure of Research
  31. Understanding Partnering Strategies in the Low-Code Platform Ecosystem
  32. Analyzing Talk and Text II: Thematic Analysis
  33. Discourse pragmatics