深度学习可解释性的相关论文
2024-06-21 20:48:25
深度学习关于模型可解释性的相关论文，及部分代码
从GitHub：awesome_deep_learning_interpretability转载
一、按时间
| Year | Publication | Paper | Citation | code |
|---|---|---|---|---|
| 2020 | CVPR | Explaining Knowledge Distillation by Quantifying the Knowledge | 3 | |
| 2020 | CVPR | High-frequency Component Helps Explain the Generalization of Convolutional Neural Networks | 16 | |
| 2020 | CVPRW | Score-CAM: Score-Weighted Visual Explanations for Convolutional Neural Networks | 7 | Pytorch |
| 2020 | ICLR | Knowledge consistency between neural networks and beyond | 3 | |
| 2020 | ICLR | Interpretable Complex-Valued Neural Networks for Privacy Protection | 2 | |
| 2019 | AI | Explanation in artificial intelligence: Insights from the social sciences | 662 | |
| 2019 | NMI | Stop Explaining Black Box Machine Learning Models for High Stakes Decisions and Use Interpretable Models Instead | 389 | |
| 2019 | NeurIPS | Can you trust your model’s uncertainty? Evaluating predictive uncertainty under dataset shift | 136 | - |
| 2019 | NeurIPS | This looks like that: deep learning for interpretable image recognition | 80 | Pytorch |
| 2019 | NeurIPS | A benchmark for interpretability methods in deep neural networks | 28 | |
| 2019 | NeurIPS | Full-gradient representation for neural network visualization | 7 | |
| 2019 | NeurIPS | On the (In) fidelity and Sensitivity of Explanations | 13 | |
| 2019 | NeurIPS | Towards Automatic Concept-based Explanations | 25 | Tensorflow |
| 2019 | NeurIPS | CXPlain: Causal explanations for model interpretation under uncertainty | 12 | |
| 2019 | CVPR | Interpreting CNNs via Decision Trees | 85 | |
| 2019 | CVPR | From Recognition to Cognition: Visual Commonsense Reasoning | 97 | Pytorch |
| 2019 | CVPR | Attention branch network: Learning of attention mechanism for visual explanation | 39 | |
| 2019 | CVPR | Interpretable and fine-grained visual explanations for convolutional neural networks | 18 | |
| 2019 | CVPR | Learning to Explain with Complemental Examples | 12 | |
| 2019 | CVPR | Revealing Scenes by Inverting Structure from Motion Reconstructions | 20 | Tensorflow |
| 2019 | CVPR | Multimodal Explanations by Predicting Counterfactuality in Videos | 4 | |
| 2019 | CVPR | Visualizing the Resilience of Deep Convolutional Network Interpretations | 1 | |
| 2019 | ICCV | U-CAM: Visual Explanation using Uncertainty based Class Activation Maps | 10 | |
| 2019 | ICCV | Towards Interpretable Face Recognition | 7 | |
| 2019 | ICCV | Taking a HINT: Leveraging Explanations to Make Vision and Language Models More Grounded | 28 | |
| 2019 | ICCV | Understanding Deep Networks via Extremal Perturbations and Smooth Masks | 17 | Pytorch |
| 2019 | ICCV | Explaining Neural Networks Semantically and Quantitatively | 6 | |
| 2019 | ICLR | Hierarchical interpretations for neural network predictions | 24 | Pytorch |
| 2019 | ICLR | How Important Is a Neuron? | 32 | |
| 2019 | ICLR | Visual Explanation by Interpretation: Improving Visual Feedback Capabilities of Deep Neural Networks | 13 | |
| 2018 | ICML | Extracting Automata from Recurrent Neural Networks Using Queries and Counterexamples | 71 | Pytorch |
| 2019 | ICML | Towards A Deep and Unified Understanding of Deep Neural Models in NLP | 15 | Pytorch |
| 2019 | ICAIS | Interpreting black box predictions using fisher kernels | 24 | |
| 2019 | ACMFAT | Explaining explanations in AI | 119 | |
| 2019 | AAAI | Interpretation of neural networks is fragile | 130 | Tensorflow |
| 2019 | AAAI | Classifier-agnostic saliency map extraction | 8 | |
| 2019 | AAAI | Can You Explain That? Lucid Explanations Help Human-AI Collaborative Image Retrieval | 1 | |
| 2019 | AAAIW | Unsupervised Learning of Neural Networks to Explain Neural Networks | 10 | |
| 2019 | AAAIW | Network Transplanting | 4 | |
| 2019 | CSUR | A Survey of Methods for Explaining Black Box Models | 655 | |
| 2019 | JVCIR | Interpretable convolutional neural networks via feedforward design | 31 | Keras |
| 2019 | ExplainAI | The (Un)reliability of saliency methods | 128 | |
| 2019 | ACL | Attention is not Explanation | 157 | |
| 2019 | EMNLP | Attention is not not Explanation | 57 | |
| 2019 | arxiv | Attention Interpretability Across NLP Tasks | 16 | |
| 2019 | arxiv | Interpretable CNNs | 2 | |
| 2018 | ICLR | Towards better understanding of gradient-based attribution methods for deep neural networks | 245 | |
| 2018 | ICLR | Learning how to explain neural networks: PatternNet and PatternAttribution | 143 | |
| 2018 | ICLR | On the importance of single directions for generalization | 134 | Pytorch |
| 2018 | ICLR | Detecting statistical interactions from neural network weights | 56 | Pytorch |
| 2018 | ICLR | Interpretable counting for visual question answering | 29 | Pytorch |
| 2018 | CVPR | Interpretable Convolutional Neural Networks | 250 | |
| 2018 | CVPR | Tell me where to look: Guided attention inference network | 134 | Chainer |
| 2018 | CVPR | Multimodal Explanations: Justifying Decisions and Pointing to the Evidence | 126 | Caffe |
| 2018 | CVPR | Transparency by design: Closing the gap between performance and interpretability in visual reasoning | 79 | Pytorch |
| 2018 | CVPR | Net2vec: Quantifying and explaining how concepts are encoded by filters in deep neural networks | 60 | |
| 2018 | CVPR | What have we learned from deep representations for action recognition? | 30 | |
| 2018 | CVPR | Learning to Act Properly: Predicting and Explaining Affordances from Images | 24 | |
| 2018 | CVPR | Teaching Categories to Human Learners with Visual Explanations | 20 | Pytorch |
| 2018 | CVPR | What do deep networks like to see? | 19 | |
| 2018 | CVPR | Interpret Neural Networks by Identifying Critical Data Routing Paths | 13 | Tensorflow |
| 2018 | ECCV | Deep clustering for unsupervised learning of visual features | 382 | Pytorch |
| 2018 | ECCV | Explainable neural computation via stack neural module networks | 55 | Tensorflow |
| 2018 | ECCV | Grounding visual explanations | 44 | |
| 2018 | ECCV | Textual explanations for self-driving vehicles | 59 | |
| 2018 | ECCV | Interpretable basis decomposition for visual explanation | 51 | Pytorch |
| 2018 | ECCV | Convnets and imagenet beyond accuracy: Understanding mistakes and uncovering biases | 36 | |
| 2018 | ECCV | Vqa-e: Explaining, elaborating, and enhancing your answers for visual questions | 20 | |
| 2018 | ECCV | Choose Your Neuron: Incorporating Domain Knowledge through Neuron-Importance | 16 | Pytorch |
| 2018 | ECCV | Diverse feature visualizations reveal invariances in early layers of deep neural networks | 9 | Tensorflow |
| 2018 | ECCV | ExplainGAN: Model Explanation via Decision Boundary Crossing Transformations | 6 | |
| 2018 | ICML | Interpretability beyond feature attribution: Quantitative testing with concept activation vectors | 214 | Tensorflow |
| 2018 | ICML | Learning to explain: An information-theoretic perspective on model interpretation | 117 | |
| 2018 | ACL | Did the Model Understand the Question? | 63 | Tensorflow |
| 2018 | FITEE | Visual interpretability for deep learning: a survey | 243 | |
| 2018 | NeurIPS | Sanity Checks for Saliency Maps | 249 | |
| 2018 | NeurIPS | Explanations based on the missing: Towards contrastive explanations with pertinent negatives | 79 | Tensorflow |
| 2018 | NeurIPS | Towards robust interpretability with self-explaining neural networks | 145 | Pytorch |
| 2018 | NeurIPS | Attacks meet interpretability: Attribute-steered detection of adversarial samples | 55 | |
| 2018 | NeurIPS | DeepPINK: reproducible feature selection in deep neural networks | 30 | Keras |
| 2018 | NeurIPS | Representer point selection for explaining deep neural networks | 30 | Tensorflow |
| 2018 | NeurIPS Workshop | Interpretable convolutional filters with sincNet | 37 | |
| 2018 | AAAI | Anchors: High-precision model-agnostic explanations | 366 | |
| 2018 | AAAI | Improving the adversarial robustness and interpretability of deep neural networks by regularizing their input gradients | 178 | Tensorflow |
| 2018 | AAAI | Deep learning for case-based reasoning through prototypes: A neural network that explains its predictions | 102 | Tensorflow |
| 2018 | AAAI | Interpreting CNN Knowledge via an Explanatory Graph | 79 | Matlab |
| 2018 | AAAI | Examining CNN Representations with respect to Dataset Bias | 37 | |
| 2018 | WACV | Grad-cam++: Generalized gradient-based visual explanations for deep convolutional networks | 174 | |
| 2018 | IJCV | Top-down neural attention by excitation backprop | 329 | |
| 2018 | TPAMI | Interpreting deep visual representations via network dissection | 87 | |
| 2018 | DSP | Methods for interpreting and understanding deep neural networks | 713 | |
| 2018 | Access | Peeking inside the black-box: A survey on Explainable Artificial Intelligence (XAI) | 390 | |
| 2018 | JAIR | Learning Explanatory Rules from Noisy Data | 155 | Tensorflow |
| 2018 | MIPRO | Explainable artificial intelligence: A survey | 108 | |
| 2018 | BMVC | Rise: Randomized input sampling for explanation of black-box models | 85 | |
| 2018 | arxiv | Distill-and-Compare: Auditing Black-Box Models Using Transparent Model Distillation | 30 | |
| 2018 | arxiv | Manipulating and measuring model interpretability | 133 | |
| 2018 | arxiv | How convolutional neural network see the world-A survey of convolutional neural network visualization methods | 45 | |
| 2018 | arxiv | Revisiting the importance of individual units in cnns via ablation | 43 | |
| 2018 | arxiv | Computationally Efficient Measures of Internal Neuron Importance | 1 | |
| 2017 | ICML | Understanding Black-box Predictions via Influence Functions | 767 | Pytorch |
| 2017 | ICML | Axiomatic attribution for deep networks | 755 | Keras |
| 2017 | ICML | Learning Important Features Through Propagating Activation Differences | 655 | |
| 2017 | ICLR | Visualizing deep neural network decisions: Prediction difference analysis | 271 | Caffe |
| 2017 | ICLR | Exploring LOTS in Deep Neural Networks | 27 | |
| 2017 | NeurIPS | A Unified Approach to Interpreting Model Predictions | 1411 | |
| 2017 | NeurIPS | Real time image saliency for black box classifiers | 161 | Pytorch |
| 2017 | NeurIPS | SVCCA: Singular Vector Canonical Correlation Analysis for Deep Learning Dynamics and Interpretability | 160 | |
| 2017 | CVPR | Mining Object Parts from CNNs via Active Question-Answering | 20 | |
| 2017 | CVPR | Network dissection: Quantifying interpretability of deep visual representations | 540 | |
| 2017 | CVPR | Improving Interpretability of Deep Neural Networks with Semantic Information | 56 | |
| 2017 | CVPR | MDNet: A Semantically and Visually Interpretable Medical Image Diagnosis Network | 129 | Torch |
| 2017 | CVPR | Making the V in VQA matter: Elevating the role of image understanding in Visual Question Answering | 582 | |
| 2017 | CVPR | Knowing when to look: Adaptive attention via a visual sentinel for image captioning | 620 | Torch |
| 2017 | CVPRW | Interpretable 3d human action analysis with temporal convolutional networks | 163 | |
| 2017 | ICCV | Grad-cam: Visual explanations from deep networks via gradient-based localization | 2444 | Pytorch |
| 2017 | ICCV | Interpretable Explanations of Black Boxes by Meaningful Perturbation | 419 | Pytorch |
| 2017 | ICCV | Interpretable Learning for Self-Driving Cars by Visualizing Causal Attention | 114 | |
| 2017 | ICCV | Understanding and comparing deep neural networks for age and gender classification | 52 | |
| 2017 | ICCV | Learning to disambiguate by asking discriminative questions | 12 | |
| 2017 | IJCAI | Right for the right reasons: Training differentiable models by constraining their explanations | 149 | |
| 2017 | IJCAI | Understanding and improving convolutional neural networks via concatenated rectified linear units | 276 | Caffe |
| 2017 | AAAI | Growing Interpretable Part Graphs on ConvNets via Multi-Shot Learning | 37 | Matlab |
| 2017 | ACL | Visualizing and Understanding Neural Machine Translation | 92 | |
| 2017 | EMNLP | A causal framework for explaining the predictions of black-box sequence-to-sequence models | 92 | |
| 2017 | CVPR Workshop | Looking under the hood: Deep neural network visualization to interpret whole-slide image analysis outcomes for colorectal polyps | 21 | |
| 2017 | survey | Interpretability of deep learning models: a survey of results | 99 | |
| 2017 | arxiv | SmoothGrad: removing noise by adding noise | 356 | |
| 2017 | arxiv | Interpretable & explorable approximations of black box models | 115 | |
| 2017 | arxiv | Distilling a neural network into a soft decision tree | 188 | Pytorch |
| 2017 | arxiv | Towards interpretable deep neural networks by leveraging adversarial examples | 54 | |
| 2017 | arxiv | Explainable artificial intelligence: Understanding, visualizing and interpreting deep learning models | 383 | |
| 2017 | arxiv | Contextual Explanation Networks | 35 | Pytorch |
| 2017 | arxiv | Challenges for transparency | 83 | |
| 2017 | ACMSOPP | Deepxplore: Automated whitebox testing of deep learning systems | 431 | |
| 2017 | CEURW | What does explainable AI really mean? A new conceptualization of perspectives | 117 | |
| 2017 | TVCG | ActiVis: Visual Exploration of Industry-Scale Deep Neural Network Models | 158 | |
| 2016 | NeurIPS | Synthesizing the preferred inputs for neurons in neural networks via deep generator networks | 321 | Caffe |
| 2016 | NeurIPS | Understanding the effective receptive field in deep convolutional neural networks | 436 | |
| 2016 | CVPR | Inverting Visual Representations with Convolutional Networks | 336 | |
| 2016 | CVPR | Visualizing and Understanding Deep Texture Representations | 98 | |
| 2016 | CVPR | Analyzing Classifiers: Fisher Vectors and Deep Neural Networks | 110 | |
| 2016 | ECCV | Generating Visual Explanations | 303 | Caffe |
| 2016 | ECCV | Design of kernels in convolutional neural networks for image classification | 14 | |
| 2016 | ICML | Understanding and improving convolutional neural networks via concatenated rectified linear units | 276 | |
| 2016 | ICML | Visualizing and comparing AlexNet and VGG using deconvolutional layers | 41 | |
| 2016 | EMNLP | Rationalizing Neural Predictions | 355 | Pytorch |
| 2016 | IJCV | Visualizing deep convolutional neural networks using natural pre-images | 281 | Matlab |
| 2016 | IJCV | Visualizing Object Detection Features | 27 | Caffe |
| 2016 | KDD | Why should i trust you?: Explaining the predictions of any classifier | 3511 | |
| 2016 | TVCG | Visualizing the hidden activity of artificial neural networks | 170 | |
| 2016 | TVCG | Towards better analysis of deep convolutional neural networks | 241 | |
| 2016 | NAACL | Visualizing and understanding neural models in nlp | 364 | Torch |
| 2016 | arxiv | Understanding neural networks through representation erasure) | 198 | |
| 2016 | arxiv | Grad-CAM: Why did you say that? | 130 | |
| 2016 | arxiv | Investigating the influence of noise and distractors on the interpretation of neural networks | 41 | |
| 2016 | arxiv | Attentive Explanations: Justifying Decisions and Pointing to the Evidence | 54 | |
| 2016 | arxiv | The Mythos of Model Interpretability | 1368 | |
| 2016 | arxiv | Multifaceted feature visualization: Uncovering the different types of features learned by each neuron in deep neural networks | 161 | |
| 2015 | ICLR | Striving for Simplicity: The All Convolutional Net | 2268 | Pytorch |
| 2015 | CVPR | Understanding deep image representations by inverting them | 1129 | Matlab |
| 2015 | ICCV | Understanding deep features with computer-generated imagery | 109 | Caffe |
| 2015 | ICML Workshop | Understanding Neural Networks Through Deep Visualization | 1216 | Tensorflow |
| 2015 | AAS | Interpretable classifiers using rules and Bayesian analysis: Building a better stroke prediction model | 385 | |
| 2014 | ECCV | Visualizing and Understanding Convolutional Networks | 9873 | Pytorch |
| 2014 | ICLR | Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps | 2745 | Pytorch |
| 2013 | ICCV | Hoggles: Visualizing object detection features | 301 |
二、按引用数量
| Year | Publication | Paper | Citation |
|---|---|---|---|
| 2014 | ECCV | Visualizing and Understanding Convolutional Networks | 8009 |
| 2016 | KDD | Why should i trust you?: Explaining the predictions of any classifier | 2255 |
| 2014 | ICLR | Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps | 2014 |
| 2015 | ICLR | Striving for Simplicity: The All Convolutional Net | 1762 |
| 2017 | ICCV | Grad-cam: Visual explanations from deep networks via gradient-based localization | 1333 |
| 2015 | ICMLW | Understanding Neural Networks Through Deep Visualization | 974 |
| 2016 | arxiv | The Mythos of Model Interpretability | 951 |
| 2015 | CVPR | Understanding deep image representations by inverting them | 929 |
| 2017 | NIPS | A Unified Approach to Interpreting Model Predictions | 591 |
| 2017 | ICML | Understanding Black-box Predictions via Influence Functions | 517 |
| 2018 | DSP | Methods for interpreting and understanding deep neural networks(scihub) | 469 |
| 2017 | CVPR | Knowing when to look: Adaptive attention via a visual sentinel for image captioning | 458 |
| 2017 | ICML | Axiomatic attribution for deep networks | 448 |
| 2017 | CVPR | Making the V in VQA matter: Elevating the role of image understanding in Visual Question Answering | 393 |
| 2017 | ICML | Learning Important Features Through Propagating Activation Differences | 383 |
| 2019 | AI | Explanation in artificial intelligence: Insights from the social sciences | 380 |
| 2017 | CVPR | Network dissection: Quantifying interpretability of deep visual representations | 373 |
| 2019 | CSUR | A Survey of Methods for Explaining Black Box Models | 344 |
| 2016 | NIPS | Understanding the effective receptive field in deep convolutional neural networks | 310 |
| 2015 | AAS | Interpretable classifiers using rules and Bayesian analysis: Building a better stroke prediction model | 304 |
| 2017 | ACMSOPP | Deepxplore: Automated whitebox testing of deep learning systems | 302 |
| 2017 | ICCV | Interpretable Explanations of Black Boxes by Meaningful Perturbation | 284 |
| 2016 | NAACL | Visualizing and understanding neural models in nlp | 269 |
| 2016 | CVPR | Inverting Visual Representations with Convolutional Networks | 266 |
| 2018 | IJCV | Top-down neural attention by excitation backprop | 256 |
| 2016 | NIPS | Synthesizing the preferred inputs for neurons in neural networks via deep generator networks | 251 |
| 2016 | EMNLP | Rationalizing Neural Predictions | 247 |
| 2016 | ECCV | Generating Visual Explanations | 224 |
| 2016 | ICML | Understanding and improving convolutional neural networks via concatenated rectified linear units | 216 |
| 2016 | IJCV | Visualizing deep convolutional neural networks using natural pre-images | 216 |
| 2017 | ICLR | Visualizing deep neural network decisions: Prediction difference analysis | 212 |
| 2017 | arxiv | SmoothGrad: removing noise by adding noise | 212 |
| 2017 | arxiv | Explainable artificial intelligence: Understanding, visualizing and interpreting deep learning models | 210 |
| 2018 | AAAI | Anchors: High-precision model-agnostic explanations | 200 |
| 2016 | TVCG | Towards better analysis of deep convolutional neural networks | 184 |
| 2018 | ECCV | Deep clustering for unsupervised learning of visual features | 167 |
| 2018 | CVPR | Interpretable Convolutional Neural Networks | 154 |
| 2018 | FITEE | Visual interpretability for deep learning: a survey | 140 |
| 2016 | arxiv | Understanding neural networks through representation erasure | 137 |
| 2018 | Access | Peeking inside the black-box: A survey on Explainable Artificial Intelligence (XAI) | 131 |
| 2016 | arxiv | Multifaceted feature visualization: Uncovering the different types of features learned by each neuron in deep neural networks | 130 |
| 2017 | arxiv | Distilling a neural network into a soft decision tree | 126 |
| 2018 | ICLR | Towards better understanding of gradient-based attribution methods for deep neural networks | 123 |
| 2018 | NIPS | Sanity Checks for Saliency Maps | 122 |
| 2016 | TVCG | Visualizing the hidden activity of artificial neural networks | 122 |
| 2017 | TVCG | ActiVis: Visual Exploration of Industry-Scale Deep Neural Network Models | 113 |
| 2018 | AAAI | Improving the adversarial robustness and interpretability of deep neural networks by regularizing their input gradients | 112 |
| 2017 | NIPS | Real time image saliency for black box classifiers | 111 |
| 2018 | ICML | Interpretability beyond feature attribution: Quantitative testing with concept activation vectors (tcav) | 110 |
| 2017 | CVPR | Interpretable 3d human action analysis with temporal convolutional networks | 106 |
| 2017 | IJCAI | Right for the right reasons: Training differentiable models by constraining their explanations | 102 |
| 2017 | NIPS | SVCCA: Singular Vector Canonical Correlation Analysis for Deep Learning Dynamics and Interpretability | 97 |
| 2019 | ExplainAI | The (Un)reliability of saliency methods(scihub) | 95 |
| 2015 | ICCV | Understanding deep features with computer-generated imagery | 94 |
| 2018 | ICLR | Learning how to explain neural networks: PatternNet and PatternAttribution | 90 |
| 2018 | JAIR | Learning Explanatory Rules from Noisy Data | 90 |
| 2016 | arxiv | Grad-CAM: Why did you say that? | 87 |
| 2017 | CVPR | MDNet: A Semantically and Visually Interpretable Medical Image Diagnosis Network | 86 |
| 2018 | WACV | Grad-cam++: Generalized gradient-based visual explanations for deep convolutional networks | 85 |
| 2016 | CVPR | Visualizing and Understanding Deep Texture Representations | 83 |
| 2016 | CVPR | Analyzing Classifiers: Fisher Vectors and Deep Neural Networks | 82 |
| 2018 | ICLR | On the importance of single directions for generalization | 81 |
| 2018 | CVPR | Tell me where to look: Guided attention inference network | 81 |
| 2017 | ICCV | Interpretable Learning for Self-Driving Cars by Visualizing Causal Attention | 80 |
| 2018 | CVPR | Multimodal Explanations: Justifying Decisions and Pointing to the Evidence | 78 |
| 2018 | arxiv | Manipulating and measuring model interpretability | 73 |
| 2018 | ICML | Learning to explain: An information-theoretic perspective on model interpretation | 72 |
| 2017 | arxiv | Challenges for transparency | 69 |
| 2017 | arxiv | Interpretable & explorable approximations of black box models | 68 |
| 2018 | AAAI | Deep learning for case-based reasoning through prototypes: A neural network that explains its predictions | 67 |
| 2017 | EMNLP | A causal framework for explaining the predictions of black-box sequence-to-sequence models | 64 |
| 2017 | CEURW | What does explainable AI really mean? A new conceptualization of perspectives | 64 |
| 2019 | AAAI | Interpretation of neural networks is fragile | 63 |
| 2019 | ACL | Attention is not Explanation | 57 |
| 2018 | TPAMI | Interpreting deep visual representations via network dissection | 56 |
| 2017 | ACL | Visualizing and Understanding Neural Machine Translation | 56 |
| 2019 | NMI | Stop Explaining Black Box Machine Learning Models for High Stakes Decisions and Use Interpretable Models Instead | 54 |
| 2019 | ACMFAT | Explaining explanations in AI | 54 |
| 2018 | CVPR | Transparency by design: Closing the gap between performance and interpretability in visual reasoning | 54 |
| 2018 | AAAI | Interpreting CNN Knowledge via an Explanatory Graph | 54 |
| 2018 | MIPRO | Explainable artificial intelligence: A survey | 54 |
| 2019 | CVPR | Interpreting CNNs via Decision Trees | 49 |
| 2017 | survey | Interpretability of deep learning models: a survey of results | 49 |
| 2018 | ICML | Extracting Automata from Recurrent Neural Networks Using Queries and Counterexamples | 47 |
| 2019 | CVPR | From Recognition to Cognition: Visual Commonsense Reasoning | 44 |
| 2017 | arxiv | Towards interpretable deep neural networks by leveraging adversarial examples | 44 |
| 2017 | CVPR | Improving Interpretability of Deep Neural Networks with Semantic Information | 43 |
| 2016 | arxiv | Attentive Explanations: Justifying Decisions and Pointing to the Evidence | 41 |
| 2018 | ECCV | Explainable neural computation via stack neural module networks | 40 |
| 2018 | CVPR | Net2vec: Quantifying and explaining how concepts are encoded by filters in deep neural networks | 39 |
| 2017 | ICCV | Understanding and comparing deep neural networks for age and gender classification | 39 |
| 2018 | ECCV | Grounding visual explanations | 38 |
| 2019 | NIPS | This looks like that: deep learning for interpretable image recognition | 35 |
| 2018 | NIPS | Explanations based on the missing: Towards contrastive explanations with pertinent negatives | 35 |
| 2017 | IJCAI | Understanding and improving convolutional neural networks via concatenated rectified linear units | 35 |
| 2018 | ACL | Did the Model Understand the Question? | 34 |
| 2018 | ICLR | Detecting statistical interactions from neural network weights | 30 |
| 2018 | ECCV | Textual explanations for self-driving vehicles | 30 |
| 2018 | BMVC | Rise: Randomized input sampling for explanation of black-box models | 30 |
| 2017 | arxiv | Contextual Explanation Networks | 28 |
| 2016 | ICML | Visualizing and comparing AlexNet and VGG using deconvolutional layers | 28 |
| 2018 | NIPS | Towards robust interpretability with self-explaining neural networks | 27 |
| 2018 | AIES | Detecting Bias in Black-Box Models Using Transparent Model Distillation | 27 |
| 2018 | arxiv | How convolutional neural network see the world-A survey of convolutional neural network visualization methods | 27 |
| 2018 | ECCV | Interpretable basis decomposition for visual explanation | 26 |
| 2018 | NIPS | Attacks meet interpretability: Attribute-steered detection of adversarial samples | 26 |
| 2017 | ICLR | Exploring LOTS in Deep Neural Networks | 26 |
| 2017 | AAAI | Growing Interpretable Part Graphs on ConvNets via Multi-Shot Learning | 26 |
| 2018 | arxiv | Revisiting the importance of individual units in cnns via ablation | 25 |
| 2018 | AAAI | Examining CNN Representations with respect to Dataset Bias | 24 |
| 2016 | arxiv | Investigating the influence of noise and distractors on the interpretation of neural networks | 24 |
| 2016 | IJCV | Visualizing Object Detection Features | 22 |
| 2018 | ICLR | Interpretable counting for visual question answering | 21 |
| 2018 | CVPR | What have we learned from deep representations for action recognition? | 20 |
| 2018 | CVPR | Learning to Act Properly: Predicting and Explaining Affordances from Images | 17 |
| 2018 | ECCV | Convnets and imagenet beyond accuracy: Understanding mistakes and uncovering biases | 17 |
| 2018 | NIPS Workshop | Interpretable Convolutional Filters with SincNet | 17 |
| 2019 | JVCIR | Interpretable convolutional neural networks via feedforward design | 16 |
| 2019 | ICLR | Hierarchical interpretations for neural network predictions | 15 |
| 2018 | NIPS | DeepPINK: reproducible feature selection in deep neural networks | 15 |
| 2017 | CVPR | Mining Object Parts from CNNs via Active Question-Answering | 15 |
| 2019 | CVPR | Attention branch network: Learning of attention mechanism for visual explanation | 14 |
| 2017 | CVPRW | Looking under the hood: Deep neural network visualization to interpret whole-slide image analysis outcomes for colorectal polyps | 14 |
| 2018 | CVPR | Teaching Categories to Human Learners with Visual Explanations | 13 |
| 2018 | ECCV | Vqa-e: Explaining, elaborating, and enhancing your answers for visual questions | 12 |
| 2018 | NIPS | Representer point selection for explaining deep neural networks | 11 |
| 2016 | ECCV | Design of kernels in convolutional neural networks for image classification | 11 |
| 2019 | ICLR | How Important Is a Neuron? | 10 |
| 2017 | ICCV | Learning to disambiguate by asking discriminative questions | 10 |
| 2019 | AAAIW | Unsupervised Learning of Neural Networks to Explain Neural Networks | 9 |
| 2018 | CVPR | What do Deep Networks Like to See? | 9 |
| 2019 | CVPR | Interpretable and fine-grained visual explanations for convolutional neural networks | 8 |
| 2018 | ECCV | Choose Your Neuron: Incorporating Domain Knowledge through Neuron-Importance | 8 |
| 2019 | ICLR | Visual Explanation by Interpretation: Improving Visual Feedback Capabilities of Deep Neural Networks | 7 |
| 2019 | ICAIS | Interpreting black box predictions using fisher kernels | 7 |
| 2019 | CVPR | Learning to Explain with Complemental Examples | 6 |
| 2019 | ICCV | U-CAM: Visual Explanation using Uncertainty based Class Activation Maps | 6 |
| 2019 | ICCV | Towards Interpretable Face Recognition | 6 |
| 2019 | CVPR | Revealing Scenes by Inverting Structure from Motion Reconstructions | 5 |
| 2019 | ICCV | Taking a HINT: Leveraging Explanations to Make Vision and Language Models More Grounded | 5 |
| 2018 | CVPR | Interpret Neural Networks by Identifying Critical Data Routing Paths | 5 |
| 2018 | ECCV | Diverse feature visualizations reveal invariances in early layers of deep neural networks | 5 |
| 2019 | ICML | Towards A Deep and Unified Understanding of Deep Neural Models in NLP | 4 |
| 2019 | AAAI | Classifier-agnostic saliency map extraction | 4 |
| 2019 | AAAIW | Network Transplanting | 4 |
| 2019 | arxiv | Attention Interpretability Across NLP Tasks | 4 |
| 2019 | NIPS | A benchmark for interpretability methods in deep neural networks（同arxiv:1806.10758） | 3 |
| 2019 | arxiv | Interpretable CNNs | 3 |
| 2019 | NIPS | Full-gradient representation for neural network visualization | 2 |
| 2019 | NIPS | On the (In) fidelity and Sensitivity of Explanations | 2 |
| 2019 | ICCV | Understanding Deep Networks via Extremal Perturbations and Smooth Masks | 2 |
| 2019 | NIPS | Towards Automatic Concept-based Explanations | 1 |
| 2019 | NIPS | CXPlain: Causal explanations for model interpretation under uncertainty | 1 |
| 2019 | CVPR | Multimodal Explanations by Predicting Counterfactuality in Videos | 1 |
| 2019 | CVPR | Visualizing the Resilience of Deep Convolutional Network Interpretations | 1 |
| 2019 | ICCV | Explaining Neural Networks Semantically and Quantitatively | 1 |
| 2018 | arxiv | Computationally Efficient Measures of Internal Neuron Importance | 1 |
| 2020 | ICLR | Knowledge Isomorphism between Neural Networks | 0 |
| 2020 | ICLR | Interpretable Complex-Valued Neural Networks for Privacy Protection | 0 |
| 2019 | AAAI | Can You Explain That? Lucid Explanations Help Human-AI Collaborative Image Retrieval | 0 |
| 2018 | ECCV | ExplainGAN: Model Explanation via Decision Boundary Crossing Transformations | 0 |
深度学习可解释性的相关论文相关推荐
- 基于深度学习实现行人跟踪相关论文总结
基于深度学习实现行人跟踪相关论文总结 [1] Recurrent YOLO and LSTM-based IR single pedestrian tracking 本文提出了一种基于空间监督的递归卷 ...
- 工程机械的深度学习和计算机视觉算法相关论文简介
目录 0.State-of-the-art review of geotechnical-driven artificial intelligence techniques in undergroun ...
- 深度学习可解释性研究（二）： Understanding Black-box Predictions via Influence Functions（详细公式推导）
该文章从鲁棒性的角度出发,对深度学习模型的可解释性进行分析,这应该是我目前为止读过最难的一篇文章,有很多公式需要推导理解,有很多细节需要慢慢品味,但文章确实是难得一见的好文章,我会尽可能把我自己的理解 ...
- 深度学习可解释性问题如何解决？图灵奖得主Bengio有一个解
作者 | Yoshua Bengio, Tristan Deleu等 译者 | 刘畅,编辑 | Just 出品 | AI科技大本营(ID:rgznai100) 自 2012 年以来,深度学习的发展有目 ...
- 阅读笔记2： 深度学习可解释性学习：不要做事后解释
选择可解释性高的机器学习模型, 而不是解释决策风险高的黑匣子模型 (原论文名:Stop Explaining Black Box Machine Learning Models for High St ...
- 深度学习数据特征提取：ICCV2019论文解析
深度学习数据特征提取:ICCV2019论文解析 Goal-Driven Sequential Data Abstraction 论文链接: http://openaccess.thecvf.com/c ...
- [论文总结] 深度学习在农业领域应用论文笔记5
深度学习在农业领域应用论文笔记5 1. Channel pruned YOLO V5s-based deep learning approach for rapid and accurate appl ...
- 国内常见与人工智能（深度学习、机器学习）相关比赛合集
国内常见与人工智能(深度学习.机器学习)相关比赛合集 以下比赛一般最近一年一次,一届一届的办,不过在我们学校都不算ABC类,但在要继续走一下还是挺有用的,当然钱也比普通比赛奖励的多 一.[第五届中 ...
- 深度学习可解释性分析-Grad-CAM
概念介绍 CAM揭示了卷积神经网络分类模型中图像的空间特征与其类别权重之间的联系,然鹅,CAM只适用于模型中有全局平均池化层并且只有一个全连接层(即输出层)的情形,如ResNet,MobileNet等 ...
最新文章
- ubuntu phpinfo mysql_Ubuntu 16.04 apache+php+mysql环境搭建
- linux实用小工具,Linux系统小工具使用大用途---1
- 【机器视觉】 endtry算子
- [转载] 杜拉拉升职记——23 “You deserve it”的两种解释
- JEECG开源团队，欢迎技术爱好者加入
- 轻量级持久存储系统 MemcacheDB
- 关于AE中出现 “对 COM 组件的调用返回了错误 HRESULT E_FAIL” 错误
- 课设题目：哈希表实现电话号码查找系统
- 基于单片机的c语言交通控制器设计论文初稿,基于AT89C51单片机的交通灯控制系统设计答辩.ppt...
- 如何将PC上的Word文档分享到朋友圈
- 判断字符串子序列【python】
- 网闸标底--网神G6150-C022
- 关于DM36x IPNC中IRCUT的使用
- SVG中插入HTML标签
- dpdk 问题分析：ice 100G 网卡 rx_packets 与 rx_bytes 统计问题
- ARM TTBR0TTBR1寄存器与ARM32页表复制。TTBR0TTBR1两寄存器在ARM,ARM32，ARM64linux中分页管理页表基地址的存放不同，CP15 CP2控制TTBR01附加协处理
- 定位教程6---上下相机
- synonyms的使用
- 复数运算（对数，指数，正弦）
- linux查看虚拟机ip的方法