Visual Genealogy of Deep Neural Networks

  title={Visual Genealogy of Deep Neural Networks},
  author={Qianwen Wang and Jun Yuan and Shuxin Chen and Hang Su and Huamin Qu and Shixia Liu},
  journal={IEEE Transactions on Visualization and Computer Graphics},
A comprehensive and comprehensible summary of existing deep neural networks (DNNs) helps practitioners understand the behaviour and evolution of DNNs, offers insights for architecture optimization, and sheds light on the working mechanisms of DNNs. However, this summary is hard to obtain because of the complexity and diversity of DNN architectures. To address this issue, we develop DNN Genealogy, an interactive visualization tool, to offer a visual summary of representative DNNs and their… Expand
A Visual Analytics Framework for Explaining and Diagnosing Transfer Learning Processes
This paper presents a visual analytics framework for the multi-level exploration of the transfer learning processes when training deep neural networks, and establishes a multi-aspect design to explain how the learned knowledge from the existing model is transferred into the new learning task when trainingDeep neural networks. Expand
OoDAnalyzer: Interactive Analysis of Out-of-Distribution Samples
This article proposes OoDAnalyzer, a visual analysis approach for interactively identifying OoD samples and explaining them in context that integrates an ensemble OoD detection method and a grid-based visualization. Expand
A Visual Designer of Layer‐wise Relevance Propagation Models
This work develops a visual model designer, named as VisLRPDesigner, to overcome the challenges in the design and use of LRP models, and facilitates relevance‐based visual analysis with two important functions: relevance‐ based pixel flipping and neuron ablation. Expand
Visual Analysis of Spatiotemporal Data Predictions with Deep Learning Models
A visualization system to analyze deep learning models with air pollution data by applying temporal prediction networks, including gated recurrent units (GRU), long short term memory (LSTM), and spatiotemporal prediction networks (convolutional LSTM). Expand
A Survey of Visual Analytics Techniques for Machine Learning
This work systematically review 259 papers published in the last ten years together with representative works before 2010 to better identify which research topics are promising and how to apply relevant techniques in visual analytics. Expand
XNLP: A Living Survey for XAI Research in Natural Language Processing
It is hoped that XNLP can become a leading demonstrative example of a living survey, balancing the depth and quality of a traditional well-constructed survey paper with the collaborative dynamism of a widely available interactive tool. Expand
Net2Vis – A Visual Grammar for Automatically Generating Publication-Tailored CNN Architecture Visualizations
This work proposes a visual grammar for convolutional neural networks (CNNs), which has been derived from an analysis of figures extracted from all ICCV and CVPR papers published between 2013 and 2019, and incorporates visual encoding, network layout, layer aggregation, and legend generation. Expand
PipelineProfiler: A Visual Analytics Tool for the Exploration of AutoML Pipelines
The Pipeline Profiler is an interactive visualization tool that allows the exploration and comparison of the solution space of machine learning pipelines produced by AutoML systems, providing users a better understanding of the algorithms that generated them as well as insights into how they can be improved. Expand
A Multidisciplinary Survey and Framework for Design and Evaluation of Explainable AI Systems
A framework with step-by-step design guidelines paired with evaluation methods to close the iterative design and evaluation cycles in multidisciplinary XAI teams is developed and summarized ready-to-use tables of evaluation methods and recommendations for different goals in XAI research are provided. Expand


Visualizing Dataflow Graphs of Deep Learning Models in TensorFlow
Overall, users find the TensorFlow Graph Visualizer useful for understanding, debugging, and sharing the structures of their models. Expand
LSTMVis: A Tool for Visual Analysis of Hidden State Dynamics in Recurrent Neural Networks
This work presents LSTMVis, a visual analysis tool for recurrent neural networks with a focus on understanding these hidden state dynamics, and describes the domain, the different stakeholders, and their goals and tasks. Expand
DeepEyes: Progressive Visual Analytics for Designing Deep Neural Networks
This paper presents DeepEyes, a Progressive Visual Analytics system that supports the design of neural networks during training, and presents novel visualizations, supporting the identification of layers that learned a stable set of patterns and, therefore, are of interest for a detailed analysis. Expand
Towards Better Analysis of Deep Convolutional Neural Networks
A hybrid visualization is developed to disclose the multiple facets of each neuron and the interactions between them and a biclustering-based edge bundling method is proposed to reduce visual clutter caused by a large number of connections between neurons. Expand
Analyzing the Training Processes of Deep Generative Models
A visual analytics approach for better understanding and diagnosing the training process of a deep generative models (DGMs) and proposes a credit assignment algorithm that indicates how other neurons contribute to the output of the neuron causing the training failure. Expand
Dual Path Networks
This work reveals the equivalence of the state-of-the-art Residual Network (ResNet) and Densely Convolutional Network (DenseNet) within the HORNN framework, and finds that ResNet enables feature re-usage while DenseNet enables new features exploration which are both important for learning good representations. Expand
Understanding Hidden Memories of Recurrent Neural Networks
This paper presents a visual analytics method for understanding and comparing RNN models for NLP tasks, and proposes a glyph-based sequence visualization based on aggregate information to analyze the behavior of an RNN’s hidden state at the sentence-level. Expand
ActiVis: Visual Exploration of Industry-Scale Deep Neural Network Models
ActiVis is developed, deployed, and iteratively improved, an interactive visualization system for interpreting large-scale deep learning models and results and can explore complex deep neural network models at both the instance-and subset-level. Expand
Interactive visualization of genealogical graphs
A graph theoretic analysis is given, which identifies why genealogical graphs can be difficult to draw, and motivates some novel graphical representations, including one based on a dual tree, a subgraph formed by the union of two trees. Expand
Language Modeling with Gated Convolutional Networks
A finite context approach through stacked convolutions, which can be more efficient since they allow parallelization over sequential tokens, is developed and is the first time a non-recurrent approach is competitive with strong recurrent models on these large scale language tasks. Expand