论文笔记 Interpreting Black-Box Classifiers Using Instance-Level Visual Explanations
2017-05-03 10:29
330 查看
Subject: Interactive Model Analysis
Target: Verify the performance of a model
Existing methods: statistical methods, in an aggregated fashion (e.g. accuracy)
Related work:
White box approach: Aiming at visualizing the internal structures of the models
Logistic Regression: transparent weighting of the features
Black box approach
Models comparison:
ModelTracker
MLCube Explorer: data cube analysis type
Contribution: a workflow and an interface
Novelty
Focus on input/output behaviour of a model (model agnostic)
Locally and globally, decisions and feature importance
Workflow:
Core of the explanation algorithm: Removing features from a vector until the predicted label changes.
User Interface of Rivelo
Limitations: works with binary classifiers and binary features
Useful Quotes: DARPA XAI program: “the effectiveness of these systems is limited by the machines current inability to explain their decisions and actions to human users [. . .] it is essential to understand, appropriately trust, and effectively manage an emerging generation of artificially intelligent machine partners"
Reference:
[1] Tamagnini, Paolo, et al. "Interpreting Black-Box Classifiers Using Instance-Level Visual Explanations." (2017).
Target: Verify the performance of a model
Existing methods: statistical methods, in an aggregated fashion (e.g. accuracy)
Related work:
White box approach: Aiming at visualizing the internal structures of the models
Logistic Regression: transparent weighting of the features
Black box approach
Models comparison:
ModelTracker
MLCube Explorer: data cube analysis type
Contribution: a workflow and an interface
Novelty
Focus on input/output behaviour of a model (model agnostic)
Locally and globally, decisions and feature importance
Workflow:
Core of the explanation algorithm: Removing features from a vector until the predicted label changes.
User Interface of Rivelo
Limitations: works with binary classifiers and binary features
Useful Quotes: DARPA XAI program: “the effectiveness of these systems is limited by the machines current inability to explain their decisions and actions to human users [. . .] it is essential to understand, appropriately trust, and effectively manage an emerging generation of artificially intelligent machine partners"
Reference:
[1] Tamagnini, Paolo, et al. "Interpreting Black-Box Classifiers Using Instance-Level Visual Explanations." (2017).
相关文章推荐
- 论文阅读-A Workflow for Visual Diagnostics of Binary Classifiers using Instance-Level Explanations
- 论文笔记:Structure Inference Net: Object Detection Using Scene-Level Context and Instance-Level Relationships
- Correlation Filter in Visual Tracking系列一:Visual Object Tracking using Adaptive Correlation Filters 论文笔记
- 【论文笔记】Mid-level Visual Element Discovery as Discriminative Mode Seeking
- Visual Object Tracking using Adaptive Correlation Filters (MOSSE)论文笔记
- Visual Object Tracking using Adaptive Correlation Filters 论文笔记
- Visual Tracking with Online Multiple Instance Learning (MIL)目标跟踪论文笔记
- Boilerplate Detection Using Shallow Text Features论文小笔记
- [论文笔记] Anatomy of a crowdsourcing platform - Using the example of microworkers.com (IMIS, 2011)
- 论文笔记(2)-Dropout-Regularization of Neural Networks using DropConnect
- [论文笔记] On Construction of Cloud IaaS for VM Live Migration Using KVM and OpenNebula (ICA3PP, 2012)
- 论文笔记之:Instance-aware Semantic Segmentation via Multi-task Network Cascades
- 深度学习论文笔记-Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition
- 论文笔记:目标追踪-CVPR2014-Adaptive Color Attributes for Real-time Visual Tracking
- 论文笔记之: Recurrent Models of Visual Attention
- 论文笔记 《Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition》
- 【论文笔记】Unsupervised Discovery of Mid-Level Discriminative Patches
- 论文笔记(1)DenseBox: Unifying Landmark Localization with End to End Object Detection
- 【论文笔记】Unsupervised Discovery of Mid-Level Discriminative Patches
- 深度学习论文笔记--Depth Map Prediction from a Single Image using a Multi-Scale Deep Network