Open AccessPosted Content
Axiomatic Attribution for Deep Networks
TL;DR: The problem of attributing the prediction of a deep network to its input features, a problem previously studied by several other works, is studied and two fundamental axioms— Sensitivity and Implementation Invariance that attribution methods ought to satisfy are identified.
read more
Abstract: We study the problem of attributing the prediction of a deep network to its input features, a problem previously studied by several other works. We identify two fundamental axioms---Sensitivity and Implementation Invariance that attribution methods ought to satisfy. We show that they are not satisfied by most known attribution methods, which we consider to be a fundamental weakness of those methods. We use the axioms to guide the design of a new attribution method called Integrated Gradients. Our method requires no modification to the original network and is extremely simple to implement; it just needs a few calls to the standard gradient operator. We apply this method to a couple of image models, a couple of text models and a chemistry model, demonstrating its ability to debug networks, to extract rules from a network, and to enable users to engage with models better.
read more
Chat with Paper
AI Agents for this Paper
Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps
Citations
Domain Knowledge Guided Deep Atrial Fibrillation Classification and Its Visual Interpretation
Xiaoyu Li,Buyue Qian,Jishang Wei,Xianli Zhang,Chen Sirui,Qinghua Zheng +5 more
- 03 Nov 2019
TL;DR: A domain knowledge guided deep neural network is proposed that uses a deep residual network as a classification framework, within which key feature ~(P-wave and R-peak position) reconstruction tasks are adopted to incorporate domain knowledge in the learning process.
12
•Posted Content
Learning Propagation Rules for Attribution Map Generation.
TL;DR: A learnable plugin module is introduced, which enables adaptive propagation rules for each pixel, to the non-linear layers during the backward pass for mask generating, and yields state-of-the-art results and gives cleaner and more visually plausible attribution maps.
12
•Posted Content
Hide-and-Seek: A Template for Explainable AI.
TL;DR: This study proposes a novel framework called Hide-and-Seek (HnS) for training Interpretable Neural Networks and establishes a theoretical foundation for exploring and comparing similar ideas.
12
•Posted Content
Explaining Bayesian Neural Networks.
Kirill Bykov,Marina M.-C. Höhne,Adelaida Creosteanu,Klaus-Robert Müller,Frederick Klauschen,Shinichi Nakajima,Marius Kloft +6 more
TL;DR: In this paper, the authors propose a holistic explanation framework for explaining BNNs, where the network weights follow a probability distribution, and thus the standard explanation extends to an explanation distribution.
12
•Posted Content
Can Explanations Be Useful for Calibrating Black Box Models
Xi Ye,Greg Durrett +1 more
TL;DR: In this article, a set of features combining human intuition about the task with model attributions generated by black box interpretation techniques are used to calibrate or rerank the model's predictions based on the features.
12
References
Going deeper with convolutions
Christian Szegedy,Wei Liu,Yangqing Jia,Pierre Sermanet,Scott Reed,Dragomir Anguelov,Dumitru Erhan,Vincent Vanhoucke,Andrew Rabinovich +8 more
- 07 Jun 2015
TL;DR: Inception as mentioned in this paper is a deep convolutional neural network architecture that achieves the new state of the art for classification and detection in the ImageNet Large-Scale Visual Recognition Challenge 2014 (ILSVRC14).
ImageNet Large Scale Visual Recognition Challenge
Olga Russakovsky,Jia Deng,Hao Su,Jonathan Krause,Sanjeev Satheesh,Sean Ma,Zhiheng Huang,Andrej Karpathy,Aditya Khosla,Michael S. Bernstein,Alexander C. Berg,Li Fei-Fei +11 more
TL;DR: The ImageNet Large Scale Visual Recognition Challenge (ILSVRC) as mentioned in this paper is a benchmark in object category classification and detection on hundreds of object categories and millions of images, which has been run annually from 2010 to present, attracting participation from more than fifty institutions.
•Posted Content
Neural Machine Translation by Jointly Learning to Align and Translate
TL;DR: In this paper, the authors propose to use a soft-searching model to find the parts of a source sentence that are relevant to predicting a target word, without having to form these parts as a hard segment explicitly.
20.9K
"Why Should I Trust You?": Explaining the Predictions of Any Classifier
Marco Tulio Ribeiro,Sameer Singh,Carlos Guestrin +2 more
- 13 Aug 2016
TL;DR: In this article, the authors propose LIME, a method to explain models by presenting representative individual predictions and their explanations in a non-redundant way, framing the task as a submodular optimization problem.
Visualizing and Understanding Convolutional Networks
Matthew D. Zeiler,Rob Fergus +1 more
- 06 Sep 2014
TL;DR: A novel visualization technique is introduced that gives insight into the function of intermediate feature layers and the operation of the classifier in large Convolutional Network models, used in a diagnostic role to find model architectures that outperform Krizhevsky et al on the ImageNet classification benchmark.
16.6K
Related Papers (5)
Matthew D. Zeiler,Rob Fergus +1 more
- 06 Sep 2014
Kaiming He,Xiangyu Zhang,Shaoqing Ren,Jian Sun +3 more
- 27 Jun 2016