Supplementary Materials for “Interpretable Explanations of Black Boxes by Meaningful Perturbation”
نویسندگان
چکیده
6. More examples of learned masks 3 6.1. Randomly selected examples . . . . . . . . . 3 6.2. Examples of the top5 hyperparameter setting 3 6.3. Examples for Different Network Architectures 3 6.4. Examples For Visualizing Different Layers . 3 6.5. Examples Comparing the Preservation vs. Deletion Loss . . . . . . . . . . . . . . . . 3 6.6. Failure Examples on Imagenet . . . . . . . . 7 6.7. Examples on COCO . . . . . . . . . . . . . 7
منابع مشابه
Local Interpretable Model-Agnostic Explanations for Music Content Analysis
The interpretability of a machine learning model is essential for gaining insight into model behaviour. While some machine learning models (e.g., decision trees) are transparent, the majority of models used today are still black-boxes. Recent work in machine learning aims to analyse these models by explaining the basis of their decisions. In this work, we extend one such technique, called local...
متن کاملInterpreting Deep Visual Representations via Network Dissection
The success of recent deep convolutional neural networks (CNNs) depends on learning hidden representations that can summarize the important factors of variation behind the data. However, CNNs often criticized as being black boxes that lack interpretability, since they have millions of unexplained model parameters. In this work, we describe Network Dissection, a method that interprets networks b...
متن کاملA Model Explanation System: Latest Updates and Extensions
We propose a general model explanation system (MES) for “explaining” the output of black box classifiers. This paper describes extensions to Turner (2015), which is referred to frequently in the text. We use the motivating example of a classifier trained to detect fraud in a credit card transaction history. The key aspect is that we provide explanations applicable to a single prediction, rather...
متن کاملHybrid Decision Making: When Interpretable Models Collaborate With Black-Box Models
Interpretable machine learning models have received increasing interest in recent years, especially in domains where humans are involved in the decision-making process. However, the possible loss of the task performance for gaining interpretability is oen inevitable. is performance downgrade puts practitioners in a dilemma of choosing between a top-performing black-box model with no explanati...
متن کاملnature|methods PhenoRipper: software for rapidly profiling microscopy images
Supplementary Figure 1 PhenoRipper flow chart. Supplementary Figure 2 Comparison of classification performance for various unsupervised image analysis methods. Supplementary Figure 3 PhenoRipper enables intuitive exploration of phenotypic space spanned by cellular perturbations. Supplementary Figure 4 Parameter dependence of PhenoRipper results. Supplementary Figure 5 PhenoRipper identifies mea...
متن کامل