Striving for Simplicity: The All Convolutional Net

Abstract

Most modern convolutional neural networks (CNNs) used for object recognition are built using the same principles: Alternating convolution and max-pooling layers followed by a small number of fully connected layers. We re-evaluate the state of the art for object recognition from small images with convolutional networks, questioning the necessity of different components in the pipeline. We find that max-pooling can simply be replaced by a convolutional layer with increased stride without loss in accuracy on several image recognition benchmarks. Following this finding -- and building on other recent work for finding simple network structures -- we propose a new architecture that consists solely of convolutional layers and yields competitive or state of the art performance on several object recognition datasets (CIFAR-10, CIFAR-100, ImageNet). To analyze the network we introduce a new variant of the "deconvolution approach" for visualizing features learned by CNNs, which can be applied to a broader range of network structures than existing approaches.

accepted to ICLR-2015 workshop track; no changes other than style

References in corpus (10)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Improving neural networks by preventing co-adaptation of feature detectors
• Caffe: Convolutional Architecture for Fast Feature Embedding
• Going Deeper with Convolutions
• Deeply-Supervised Nets
• Stochastic Pooling for Regularization of Deep Convolutional Neural Networks
• Fractional Max-Pooling
• High-Performance Neural Networks for Visual Object Classification
• Deep Networks with Internal Selective Attention through Feedback Connections
• Improving Deep Neural Networks with Probabilistic Maxout Units

Cited by in corpus (508)

• A Survey on Deep Learning in Medical Image Analysis
• Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
• Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization
• Neural Architecture Search with Reinforcement Learning
• Geometric deep learning: going beyond Euclidean data
• Efficient Multi-Scale 3D CNN with Fully Connected CRF for Accurate Brain Lesion Segmentation
• Grad-CAM++: Improved Visual Explanations for Deep Convolutional Networks
• A systematic study of the class imbalance problem in convolutional neural networks
• Axiomatic Attribution for Deep Networks
• Methods for Interpreting and Understanding Deep Neural Networks
• A Survey on Explainable Artificial Intelligence (XAI): Towards Medical XAI
• Learning Important Features Through Propagating Activation Differences
• Fast and Accurate Deep Network Learning by Exponential Linear Units (ELUs)
• Interpretable machine learning: definitions, methods, and applications
• Densely Connected Convolutional Networks
• Temporal Ensembling for Semi-Supervised Learning
• Explaining Deep Neural Networks and Beyond: A Review of Methods and Applications
• Automatically designing CNN architectures using genetic algorithm for image classification
• A Survey of Model Compression and Acceleration for Deep Neural Networks
• Medical Image Segmentation Using Deep Learning: A Survey
• Convolutional Networks for Fast, Energy-Efficient Neuromorphic Computing
• CA-Net: Comprehensive Attention Convolutional Neural Networks for Explainable Medical Image Segmentation
• Resnet in Resnet: Generalizing Residual Architectures
• Quantifying the effects of data augmentation and stain color normalization in convolutional neural networks for computational pathology
• A deep learning architecture for temporal sleep stage classification using multivariate and multimodal time series
• Semi-Supervised Learning with Ladder Networks
• Geometric Deep Learning: Grids, Groups, Graphs, Geodesics, and Gauges
• This Looks Like That: Deep Learning for Interpretable Image Recognition
• Not Just a Black Box: Learning Important Features Through Propagating Activation Differences
• Large-Scale Evolution of Image Classifiers
• Eigen-CAM: Class Activation Map using Principal Components
• Data Augmentation using Random Image Cropping and Patching for Deep CNNs
• Classifying and Segmenting Microscopy Images Using Convolutional Multiple Instance Learning
• Hyperspectral Image Classification-Traditional to Deep Models: A Survey for Future Prospects
• The History Began from AlexNet: A Comprehensive Survey on Deep Learning Approaches
• Residual Networks of Residual Networks: Multilevel Residual Networks
• Explanations in Autonomous Driving: A Survey
• Neural Encoding and Decoding with Deep Learning for Dynamic Natural Vision
• Grad-CAM: Why did you say that?
• Systematic evaluation of CNN advances on the ImageNet
• Layer-Wise Relevance Propagation for Explaining Deep Neural Network Decisions in MRI-Based Alzheimer's Disease Classification
• Synaptic Plasticity Dynamics for Deep Continuous Local Learning (DECOLLE)
• Deep Networks with Stochastic Depth
• Explainable AI for Trees: From Local Explanations to Global Understanding
• Graph-Based Deep Learning for Medical Diagnosis and Analysis: Past, Present and Future
• A Review on Generative Adversarial Networks: Algorithms, Theory, and Applications
• Automatic calcium scoring in low-dose chest CT using deep neural networks with dilated convolutions
• Conditional Generative Adversarial Networks for Speech Enhancement and Noise-Robust Speaker Verification
• Traffic Prediction using Artificial Intelligence: Review of Recent Advances and Emerging Opportunities
• Robust Large Margin Deep Neural Networks
• ReNet: A Recurrent Neural Network Based Alternative to Convolutional Networks
• Stacked What-Where Auto-encoders
• Explainable Artificial Intelligence: A Survey of Needs, Techniques, Applications, and Future Direction
• Adversarial Examples: Attacks and Defenses for Deep Learning
• Unveiling COVID-19 from Chest X-ray with deep learning: a hurdles race with small data
• A feature agnostic approach for glaucoma detection in OCT volumes
• MagNet: a Two-Pronged Defense against Adversarial Examples
• Do Convolutional Neural Networks Learn Class Hierarchy?
• Quantum Entanglement in Deep Learning Architectures
• Identity Mappings in Deep Residual Networks
• An Explainable 3D Residual Self-Attention Deep Neural Network FOR Joint Atrophy Localization and Alzheimer's Disease Diagnosis using Structural MRI
• All you need is a good init
• LipNet: End-to-End Sentence-level Lipreading
• On Empirical Comparisons of Optimizers for Deep Learning
• Guidelines and Evaluation of Clinical Explainable AI in Medical Image Analysis
• Uncertainty and Interpretability in Convolutional Neural Networks for Semantic Segmentation of Colorectal Polyps
• Sparse Networks from Scratch: Faster Training without Losing Performance
• Variational Encoding of Complex Dynamics
• Bridging the Gaps Between Residual Learning, Recurrent Neural Networks and Visual Cortex
• Adversarial Example Detection for DNN Models: A Review and Experimental Comparison
• Ground Truth Evaluation of Neural Network Explanations with CLEVR-XAI
• Understanding and Improving Convolutional Neural Networks via Concatenated Rectified Linear Units
• Ensemble of Deep Convolutional Neural Networks for Automatic Pavement Crack Detection and Measurement
• Smooth Grad-CAM++: An Enhanced Inference Level Visualization Technique for Deep Convolutional Neural Network Models
• Zero-Shot Machine Unlearning
• XCM: An Explainable Convolutional Neural Network for Multivariate Time Series Classification
• StressGAN: A Generative Deep Learning Model for 2D Stress Distribution Prediction
• CNN-based Segmentation of Medical Imaging Data
• IMAC: In-memory multi-bit Multiplication andACcumulation in 6T SRAM Array
• Further advantages of data augmentation on convolutional neural networks
• Deep Residual Networks with Exponential Linear Unit
• Training Confidence-calibrated Classifiers for Detecting Out-of-Distribution Samples
• Toward Explainable AI for Regression Models
• Recent Advances in Object Detection in the Age of Deep Convolutional Neural Networks
• A survey on GANs for computer vision: Recent research, analysis and taxonomy
• Robust Explainability: A Tutorial on Gradient-Based Attribution Methods for Deep Neural Networks
• B-CNN: Branch Convolutional Neural Network for Hierarchical Classification
• Wavelet Convolutional Neural Networks
• Review: Deep Learning in Electron Microscopy
• Advances of Machine Learning in Materials Science: Ideas and Techniques
• Neural Network Attribution Methods for Problems in Geoscience: A Novel Synthetic Benchmark Dataset
• Learning to Explain: An Information-Theoretic Perspective on Model Interpretation
• Entropy-SGD: Biasing Gradient Descent Into Wide Valleys
• Towards Explaining Anomalies: A Deep Taylor Decomposition of One-Class Models
• Decoding and mapping task states of the human brain via deep learning
• Theory and Tools for the Conversion of Analog to Spiking Convolutional Neural Networks
• Wavelet Convolutional Neural Networks for Texture Classification
• Detecting hidden signs of diabetes in external eye photographs
• ISeeU: Visually interpretable deep learning for mortality prediction inside the ICU
• Hierarchical interpretations for neural network predictions
• A Deep Learning Perspective on the Origin of Facial Expressions
• CondenseNet: An Efficient DenseNet using Learned Group Convolutions
• Score-CAM: Score-Weighted Visual Explanations for Convolutional Neural Networks
• Unsupervised Learning of View-invariant Action Representations
• Towards Interpretable Deep Neural Networks by Leveraging Adversarial Examples
• Automatic brain tumor grading from MRI data using convolutional neural networks and quality assessment
• Benchmarking Attribution Methods with Relative Feature Importance
• Scalable Gradient-Based Tuning of Continuous Regularization Hyperparameters
• Generative Adversarial Trainer: Defense to Adversarial Perturbations with GAN
• Deep Koalarization: Image Colorization using CNNs and Inception-ResNet-v2
• Deep Divergence-Based Approach to Clustering
• Discovering Parametric Activation Functions
• Interleaved Group Convolutions for Deep Neural Networks
• Gradients of Counterfactuals
• Real Time Image Saliency for Black Box Classifiers
• Scattering Networks for Hybrid Representation Learning
• Enhanced Convolutional Neural Tangent Kernels
• Explainable Artificial Intelligence for Process Mining: A General Overview and Application of a Novel Local Explanation Approach for Predictive Process Monitoring
• Modulation spectral features for speech emotion recognition using deep neural networks
• Predicting post-operative right ventricular failure using video-based deep learning
• How Important Is a Neuron?
• Counterfactual State Explanations for Reinforcement Learning Agents via Generative Deep Learning
• Do Explanations Reflect Decisions? A Machine-centric Strategy to Quantify the Performance of Explainability Algorithms
• Stacking-Based Deep Neural Network: Deep Analytic Network for Pattern Classification
• On the Potential of Simple Framewise Approaches to Piano Transcription
• NeuronInspect: Detecting Backdoors in Neural Networks via Output Explanations
• Evaluating Explainable AI on a Multi-Modal Medical Imaging Task: Can Existing Algorithms Fulfill Clinical Requirements?
• Selective Feature Connection Mechanism: Concatenating Multi-layer CNN Features with a Feature Selector
• Critical Learning Periods in Deep Neural Networks
• Real-time Convolutional Neural Networks for Emotion and Gender Classification
• Convolutional Neural Networks with Gated Recurrent Connections
• Six Degree-of-Freedom Body-Fixed Hovering over Unmapped Asteroids via LIDAR Altimetry and Reinforcement Meta-Learning
• $\mathcal{G}$-softmax: Improving Intra-class Compactness and Inter-class Separability of Features
• Comparison of Batch Normalization and Weight Normalization Algorithms for the Large-scale Image Classification
• Artificial Intelligence in Glioma Imaging: Challenges and Advances
• Continuous Human Action Recognition for Human-Machine Interaction: A Review
• Lets keep it simple, Using simple architectures to outperform deeper and more complex architectures
• A Theoretical Explanation for Perplexing Behaviors of Backpropagation-based Visualizations
• A Survey of Safety and Trustworthiness of Deep Neural Networks: Verification, Testing, Adversarial Attack and Defence, and Interpretability
• Restricting the Flow: Information Bottlenecks for Attribution
• A Spatiotemporal Deep Neural Network for Fine-Grained Multi-Horizon Wind Prediction
• TCGAN: Convolutional Generative Adversarial Network for Time Series Classification and Clustering
• A Taxonomy and Library for Visualizing Learned Features in Convolutional Neural Networks
• Forecasting Megaelectron-Volt Electrons inside Earth's Outer Radiation Belt: PreMevE 2.0 Based on Supervised Machine Learning Algorithms
• Inspect, Understand, Overcome: A Survey of Practical Methods for AI Safety
• GeneGAN: Learning Object Transfiguration and Attribute Subspace from Unpaired Data
• Uncertainty-Aware Deep Ensembles for Reliable and Explainable Predictions of Clinical Time Series
• Attention-based Convolutional Autoencoders for 3D-Variational Data Assimilation
• Compiling Spiking Neural Networks to Neuromorphic Hardware
• Deep Convolutional Neural Network Design Patterns
• On Interpretability of Artificial Neural Networks: A Survey
• Visualizing Deep Networks by Optimizing with Integrated Gradients
• Where is the Information in a Deep Neural Network?
• Computer-Aided Diagnosis of Label-Free 3-D Optical Coherence Microscopy Images of Human Cervical Tissue
• Feedback alignment in deep convolutional networks
• Parkinson's Disease Recognition Using SPECT Image and Interpretable AI: A Tutorial
• Application of Multi-channel 3D-cube Successive Convolution Network for Convective Storm Nowcasting
• Separating the EoR Signal with a Convolutional Denoising Autoencoder: A Deep-learning-based Method
• Learning Regional Attraction for Line Segment Detection
• Understanding the Disharmony between Dropout and Batch Normalization by Variance Shift
• Improving Efficiency in Convolutional Neural Network with Multilinear Filters
• Densely Connected CNNs for Bird Audio Detection
• Multi-Residual Networks: Improving the Speed and Accuracy of Residual Networks
• Towards Better Understanding Attribution Methods
• Multilinear Compressive Learning
• Resource-aware Time Series Imaging Classification for Wireless Link Layer Anomalies
• Convolutional Neural Networks Analyzed via Convolutional Sparse Coding
• Iterative Augmentation of Visual Evidence for Weakly-Supervised Lesion Localization in Deep Interpretability Frameworks: Application to Color Fundus Images
• Interpretations are useful: penalizing explanations to align neural networks with prior knowledge
• Supporting DNN Safety Analysis and Retraining through Heatmap-based Unsupervised Learning
• Deep Convolutional Neural Networks with Merge-and-Run Mappings
• A Learning Framework for n-bit Quantized Neural Networks toward FPGAs
• A Deep Learning Approach for Pose Estimation from Volumetric OCT Data
• Finding Competitive Network Architectures Within a Day Using UCT
• Rethinking Self-driving: Multi-task Knowledge for Better Generalization and Accident Explanation Ability
• An Analysis of Pre-Training on Object Detection
• Explain Your Move: Understanding Agent Actions Using Specific and Relevant Feature Attribution
• sMRI-PatchNet: A novel explainable patch-based deep learning network for Alzheimer's disease diagnosis and discriminative atrophy localisation with Structural MRI
• Exploiting saliency for object segmentation from image level labels
• Unsupervised Learning of Long-Term Motion Dynamics for Videos
• A Survey on Neural Network Interpretability
• Understanding Integrated Gradients with SmoothTaylor for Deep Neural Network Attribution
• Deep Learning for Real-Time Aerodynamic Evaluations of Arbitrary Vehicle Shapes
• Attentive Group Equivariant Convolutional Networks
• A Novel Weight-Shared Multi-Stage CNN for Scale Robustness
• White Box Methods for Explanations of Convolutional Neural Networks in Image Classification Tasks
• Adversarial Distillation of Bayesian Neural Network Posteriors
• Shapley Flow: A Graph-based Approach to Interpreting Model Predictions
• Deep Pyramidal Residual Networks
• When and How to Fool Explainable Models (and Humans) with Adversarial Examples
• Towards Best Practice of Interpreting Deep Learning Models for EEG-based Brain Computer Interfaces
• Explaining deep learning of galaxy morphology with saliency mapping
• Multi-site fMRI Analysis Using Privacy-preserving Federated Learning and Domain Adaptation: ABIDE Results
• IGCV3: Interleaved Low-Rank Group Convolutions for Efficient Deep Neural Networks
• Robust Classification with Convolutional Prototype Learning
• DeepMRSeg: A convolutional deep neural network for anatomy and abnormality segmentation on MR images
• What Do You See? Evaluation of Explainable Artificial Intelligence (XAI) Interpretability through Neural Backdoors
• Geo-Context Aware Study of Vision-Based Autonomous Driving Models and Spatial Video Data
• CNN Attention Guidance for Improved Orthopedics Radiographic Fracture Classification
• Image Classification with Hierarchical Multigraph Networks
• Smoothed Geometry for Robust Attribution
• HYDRA: Hypergradient Data Relevance Analysis for Interpreting Deep Neural Networks
• Attention Based Glaucoma Detection: A Large-scale Database and CNN Model
• Interpretable deep learning for nuclear deformation in heavy ion collisions
• Self-Explaining Structures Improve NLP Models
• DeepTracker: Visualizing the Training Process of Convolutional Neural Networks
• Spectral Multigraph Networks for Discovering and Fusing Relationships in Molecules
• Identification of Lensed Gravitational Waves with Deep Learning
• The RFML Ecosystem: A Look at the Unique Challenges of Applying Deep Learning to Radio Frequency Applications
• Learning across scales - A multiscale method for Convolution Neural Networks
• Quantitative Phase Imaging and Artificial Intelligence: A Review
• Review On Deep Learning Technique For Underwater Object Detection
• Automatic Photo Orientation Detection with Convolutional Neural Networks
• Relevance Prediction from Eye-movements Using Semi-interpretable Convolutional Neural Networks
• Wide Inference Network for Image Denoising via Learning Pixel-distribution Prior
• Exploring the Function Space of Deep-Learning Machines
• Learning Kernel for Conditional Moment-Matching Discrepancy-based Image Classification
• Deep Convolutional Neural Networks for Smile Recognition
• Generalized Octave Convolutions for Learned Multi-Frequency Image Compression
• Practical Block-wise Neural Network Architecture Generation
• Time Matters in Regularizing Deep Networks: Weight Decay and Data Augmentation Affect Early Learning Dynamics, Matter Little Near Convergence
• M3d-CAM: A PyTorch library to generate 3D data attention maps for medical deep learning
• DeepLofargram: A Deep Learning based Fluctuating Dim Frequency Line Detection and Recovery
• I-SPLIT: Deep Network Interpretability for Split Computing
• Centroid Transformers: Learning to Abstract with Attention
• Combining Strategic Learning and Tactical Search in Real-Time Strategy Games
• Parle: parallelizing stochastic gradient descent
• Explainable Differential Privacy-Hyperdimensional Computing for Balancing Privacy and Transparency in Additive Manufacturing Monitoring
• Scaling the Scattering Transform: Deep Hybrid Networks
• Take it in your stride: Do we need striding in CNNs?
• Precomputed Real-Time Texture Synthesis with Markovian Generative Adversarial Networks
• DISCOVER: 2-D Multiview Summarization of Optical Coherence Tomography Angiography for Automatic Diabetic Retinopathy Diagnosis
• Where is the Model Looking At?--Concentrate and Explain the Network Attention
• Towards Score Following in Sheet Music Images
• Learning to decompose for object detection and instance segmentation
• Classifier-agnostic saliency map extraction
• Automated Strabismus Detection for Telemedicine Applications
• Finding Better Topologies for Deep Convolutional Neural Networks by Evolution
• ResNet or DenseNet? Introducing Dense Shortcuts to ResNet
• A Biologically Plausible Learning Rule for Deep Learning in the Brain
• Co-Attentive Equivariant Neural Networks: Focusing Equivariance On Transformations Co-Occurring In Data
• Quantization of Deep Neural Networks for Accumulator-constrained Processors
• Adversarial Infidelity Learning for Model Interpretation
• FrankenSplit: Efficient Neural Feature Compression with Shallow Variational Bottleneck Injection for Mobile Edge Computing
• Knowledge Projection for Deep Neural Networks
• Energy Efficient Hadamard Neural Networks
• Deep learning for photoacoustic imaging: a survey
• Towards Robust, Locally Linear Deep Networks
• Learning Pixel-Distribution Prior with Wider Convolution for Image Denoising
• Confidential Inference via Ternary Model Partitioning
• Data Dropout: Optimizing Training Data for Convolutional Neural Networks
• Information Dropout: Learning Optimal Representations Through Noisy Computation
• Relative Attributing Propagation: Interpreting the Comparative Contributions of Individual Units in Deep Neural Networks
• Optimization of Convolutional Neural Network using Microcanonical Annealing Algorithm
• Approximating the Ideal Observer for joint signal detection and localization tasks by use of supervised learning methods
• A Peg-in-hole Task Strategy for Holes in Concrete
• Quantitative Evaluations on Saliency Methods: An Experimental Study
• Quicksilver: Fast Predictive Image Registration - a Deep Learning Approach
• Visual Explanation for Deep Metric Learning
• Techniques for visualizing LSTMs applied to electrocardiograms
• An Introduction to Deep Visual Explanation
• Dropout with Expectation-linear Regularization
• Age Group and Gender Estimation in the Wild with Deep RoR Architecture
• A Gradient Mapping Guided Explainable Deep Neural Network for Extracapsular Extension Identification in 3D Head and Neck Cancer Computed Tomography Images
• A Data-driven Neural Network Architecture for Sentiment Analysis
• Causal Interpretability for Machine Learning -- Problems, Methods and Evaluation
• Benchmarking the Influence of Pre-training on Explanation Performance in MR Image Classification
• FUTURE-AI: Guiding Principles and Consensus Recommendations for Trustworthy Artificial Intelligence in Medical Imaging
• Interpretability is a Kind of Safety: An Interpreter-based Ensemble for Adversary Defense
• Pulsar Candidate Identification with Artificial Intelligence Techniques
• Fix your classifier: the marginal value of training the last weight layer
• Rethinking the backbone architecture for tiny object detection
• One-vs-Rest Network-based Deep Probability Model for Open Set Recognition
• Predicting Peak Stresses In Microstructured Materials Using Convolutional Encoder-Decoder Learning
• Rad4XCNN: a new agnostic method for post-hoc global explanation of CNN-derived features by means of radiomics
• Convolutional Residual Memory Networks
• Improving Deep Neural Network with Multiple Parametric Exponential Linear Units
• Revealing Hidden Potentials of the q-Space Signal in Breast Cancer
• Parser-Free Virtual Try-on via Distilling Appearance Flows
• DVOLVER: Efficient Pareto-Optimal Neural Network Architecture Search
• Genre-Agnostic Key Classification With Convolutional Neural Networks
• Interpretable Spatio-temporal Attention for Video Action Recognition
• Pair the Dots: Jointly Examining Training History and Test Stimuli for Model Interpretability
• Semiotic Aggregation in Deep Learning
• A Note about: Local Explanation Methods for Deep Neural Networks lack Sensitivity to Parameter Values
• DMC-Net: Generating Discriminative Motion Cues for Fast Compressed Video Action Recognition
• Attention-based Wav2Text with Feature Transfer Learning
• Spatio-Temporal Perturbations for Video Attribution
• Sampled Training and Node Inheritance for Fast Evolutionary Neural Architecture Search
• LioNets: A Neural-Specific Local Interpretation Technique Exploiting Penultimate Layer Information
• Convolutional Networks with Dense Connectivity
• Urban morphology meets deep learning: Exploring urban forms in one million cities, town and villages across the planet
• Deep Learning for Weakly-Supervised Object Detection and Object Localization: A Survey
• Rationalizing Predictions by Adversarial Information Calibration
• Improving Neural Architecture Search Image Classifiers via Ensemble Learning
• Self-explanatory Deep Salient Object Detection
• Saliency-driven Word Alignment Interpretation for Neural Machine Translation
• Understanding Individual Decisions of CNNs via Contrastive Backpropagation
• Spatial-Winograd Pruning Enabling Sparse Winograd Convolution
• Adversarial Robustness Assessment: Why both $L_0$ and $L_{\infty }$ Attacks Are Necessary
• How Do Neural Networks Estimate Optical Flow? A Neuropsychology-Inspired Study
• Finding and Visualizing Weaknesses of Deep Reinforcement Learning Agents
• Remote Sensor Design for Visual Recognition with Convolutional Neural Networks
• Fitted Learning: Models with Awareness of their Limits
• Targeted Nonlinear Adversarial Perturbations in Images and Videos
• Interpretable multimodal fusion networks reveal mechanisms of brain cognition
• Automatic Configuration of Deep Neural Networks with EGO
• A novel method for extracting interpretable knowledge from a spiking neural classifier with time-varying synaptic weights
• PointMask: Towards Interpretable and Bias-Resilient Point Cloud Processing
• Regularizing Reasons for Outfit Evaluation with Gradient Penalty
• Deep Transform and Metric Learning Network: Wedding Deep Dictionary Learning and Neural Networks
• Improving Deep Image Clustering With Spatial Transformer Layers
• Robustness of Visual Explanations to Common Data Augmentation
• The Shapley Taylor Interaction Index
• BSED: Baseline Shapley-Based Explainable Detector
• MeLIME: Meaningful Local Explanation for Machine Learning Models
• CNNPruner: Pruning Convolutional Neural Networks with Visual Analytics
• Interpretation of Deep Temporal Representations by Selective Visualization of Internally Activated Nodes
• Propagating Asymptotic-Estimated Gradients for Low Bitwidth Quantized Neural Networks
• Out-of-Distribution Detection Using Neural Rendering Generative Models
• Transfer entropy-based feedback improves performance in artificial neural networks
• Deep Learning in Information Security
• PointCloud Saliency Maps
• Stochastic Gradient Descent with Polyak's Learning Rate
• Achievements and Challenges in Explaining Deep Learning based Computer-Aided Diagnosis Systems
• Better the Devil you Know: An Analysis of Evasion Attacks using Out-of-Distribution Adversarial Examples
• Towards a Unified Evaluation of Explanation Methods without Ground Truth
• Multiscale Hierarchical Convolutional Networks
• EvalAttAI: A Holistic Approach to Evaluating Attribution Maps in Robust and Non-Robust Models
• Information-Theoretic Visual Explanation for Black-Box Classifiers
• XLVIN: eXecuted Latent Value Iteration Nets
• Faster Convergence in Deep-Predictive-Coding Networks to Learn Deeper Representations
• Robust Models Are More Interpretable Because Attributions Look Normal
• Multi Layer Neural Networks as Replacement for Pooling Operations
• Towards Interrogating Discriminative Machine Learning Models
• Bayesian Conditional Generative Adverserial Networks
• Object Based Attention Through Internal Gating
• Improving Object Detection with Inverted Attention
• Towards Interpretable Attention Networks for Cervical Cancer Analysis
• SISC: End-to-end Interpretable Discovery Radiomics-Driven Lung Cancer Prediction via Stacked Interpretable Sequencing Cells
• A Forward-Backward Approach for Visualizing Information Flow in Deep Networks
• Interpreting Vision and Language Generative Models with Semantic Visual Priors
• Respond-CAM: Analyzing Deep Models for 3D Imaging Data by Visualizations
• Convolution Aware Initialization
• Towards Explaining Satellite Based Poverty Predictions with Convolutional Neural Networks
• CAMERAS: Enhanced Resolution And Sanity preserving Class Activation Mapping for image saliency
• Tackling extreme urban heat: a machine learning approach to assess the impacts of climate change and the efficacy of climate adaptation strategies in urban microclimates
• Maximally Invariant Data Perturbation as Explanation
• Infinite Variational Autoencoder for Semi-Supervised Learning
• NatCSNN: A Convolutional Spiking Neural Network for recognition of objects extracted from natural images
• Solving the enigma: Enhancing faithfulness and comprehensibility in explanations of deep networks
• A Committee of Convolutional Neural Networks for Image Classication in the Concurrent Presence of Feature and Label Noise
• Generalizing Deep Models for Overhead Image Segmentation Through Getis-Ord Gi* Pooling
• On The Coherence of Quantitative Evaluation of Visual Explanations
• Channel Locality Block: A Variant of Squeeze-and-Excitation
• The Resistance to Label Noise in K-NN and DNN Depends on its Concentration
• Optimising for Interpretability: Convolutional Dynamic Alignment Networks
• Patchnet: Interpretable Neural Networks for Image Classification
• Interpreting Interpretations: Organizing Attribution Methods by Criteria
• Mixture Proportion Estimation and PU Learning: A Modern Approach
• Manifold Adversarial Learning
• Explaining Motion Relevance for Activity Recognition in Video Deep Learning Models
• Fast Predictive Multimodal Image Registration
• Post-synaptic potential regularization has potential
• Challenges for cognitive decoding using deep learning methods
• Analyzing Effects of Mixed Sample Data Augmentation on Model Interpretability
• A Review of Explainable Artificial Intelligence in Manufacturing
• Inducing Causal Structure for Interpretable Neural Networks
• Skin disease diagnosis with deep learning: a review
• An Empirical Study of Explainable AI Techniques on Deep Learning Models For Time Series Tasks
• Epigenetic evolution of deep convolutional models
• TSInsight: A local-global attribution framework for interpretability in time-series data
• Do We Need Fully Connected Output Layers in Convolutional Networks?
• SegNBDT: Visual Decision Rules for Segmentation
• A psychophysics approach for quantitative comparison of interpretable computer vision models
• Differential Diagnosis for Pancreatic Cysts in CT Scans Using Densely-Connected Convolutional Networks
• Unsupervised Learning using Pretrained CNN and Associative Memory Bank
• Understanding Convolutional Networks with APPLE : Automatic Patch Pattern Labeling for Explanation
• DelugeNets: Deep Networks with Efficient and Flexible Cross-layer Information Inflows
• Tagger: Deep Unsupervised Perceptual Grouping
• Future Urban Scenes Generation Through Vehicles Synthesis
• Production-Level Facial Performance Capture Using Deep Convolutional Neural Networks
• A Singular Value Perspective on Model Robustness
• Gradually Updated Neural Networks for Large-Scale Image Recognition
• Learning Less-Overlapping Representations
• Using KL-divergence to focus Deep Visual Explanation
• Achieving Explainability for Plant Disease Classification with Disentangled Variational Autoencoders
• Deep Visual City Recognition Visualization
• Gradient Weighted Superpixels for Interpretability in CNNs
• Context Augmentation for Convolutional Neural Networks
• Visual Confusion Label Tree For Image Classification
• Attacking Convolutional Neural Network using Differential Evolution
• Detecting Adversarial Perturbations with Saliency
• Local Concept Embeddings for Analysis of Concept Distributions in Vision DNN Feature Spaces
• Embedded Encoder-Decoder in Convolutional Networks Towards Explainable AI
• Improving training of deep neural networks via Singular Value Bounding
• Fixed smooth convolutional layer for avoiding checkerboard artifacts in CNNs
• Where is the Fake? Patch-Wise Supervised GANs for Texture Inpainting
• Building a Regular Decision Boundary with Deep Networks
• Fine-grained Uncertainty Modeling in Neural Networks
• Three-body renormalization group limit cycles based on unsupervised feature learning
• Top-down Neural Attention by Excitation Backprop
• A Volumetric Convolutional Neural Network for Brain Tumor Segmentation
• Optimistic and Pessimistic Neural Networks for Scene and Object Recognition
• Finding New Diagnostic Information for Detecting Glaucoma using Neural Networks
• CrescendoNet: A Simple Deep Convolutional Neural Network with Ensemble Behavior
• Fine-Grained Energy and Performance Profiling framework for Deep Convolutional Neural Networks
• The Helmholtz Method: Using Perceptual Compression to Reduce Machine Learning Complexity
• Inherent Weight Normalization in Stochastic Neural Networks
• Farmland Parcel Delineation Using Spatio-temporal Convolutional Networks
• Reducing over-smoothness in speech synthesis using Generative Adversarial Networks
• Ontology-based Interpretable Machine Learning for Textual Data
• SIM2REALVIZ: Visualizing the Sim2Real Gap in Robot Ego-Pose Estimation
• Quantitative Parametric Mapping of Tissues Properties from Standard Magnetic Resonance Imaging Enabled by Deep Learning
• Model Interpretability and Rationale Extraction by Input Mask Optimization
• Supervised Deep Sparse Coding Networks
• EXplainable Neural-Symbolic Learning (X-NeSyL) methodology to fuse deep learning representations with expert knowledge graphs: the MonuMAI cultural heritage use case
• Finding Discriminative Filters for Specific Degradations in Blind Super-Resolution
• ExAD: An Ensemble Approach for Explanation-based Adversarial Detection
• XAI for Self-supervised Clustering of Wireless Spectrum Activity
• Brand > Logo: Visual Analysis of Fashion Brands
• Toward the Understanding of Deep Text Matching Models for Information Retrieval
• Semantic-guided modeling of spatial relation and object co-occurrence for indoor scene recognition
• Adversarial TCAV -- Robust and Effective Interpretation of Intermediate Layers in Neural Networks
• Contextual Local Explanation for Black Box Classifiers
• FedGPO: Heterogeneity-Aware Global Parameter Optimization for Efficient Federated Learning
• Regional Image Perturbation Reduces $L_p$ Norms of Adversarial Examples While Maintaining Model-to-model Transferability
• ImageNet-trained deep neural network exhibits illusion-like response to the Scintillating Grid
• Don't Fear Peculiar Activation Functions: EUAF and Beyond
• Crowdsourcing Evaluation of Saliency-based XAI Methods
• First-Order Preconditioning via Hypergradient Descent
• Learning Preference-Based Similarities from Face Images using Siamese Multi-Task CNNs
• Fast and interpretable classification of small X-ray diffraction datasets using data augmentation and deep neural networks
• Interpretable Disentanglement of Neural Networks by Extracting Class-Specific Subnetwork
• Sample Variance Decay in Randomly Initialized ReLU Networks
• Pattern-Guided Integrated Gradients
• Angular Learning: Toward Discriminative Embedded Features
• Understanding Patch-Based Learning by Explaining Predictions
• Weakly-Supervised Cell Tracking via Backward-and-Forward Propagation
• Backdrop: Stochastic Backpropagation
• Black-box Adversarial Sample Generation Based on Differential Evolution
• Deep learning improved by biological activation functions
• Understanding Regularization to Visualize Convolutional Neural Networks
• Transparency guided ensemble convolutional neural networks for stratification of pseudoprogression and true progression of glioblastoma multiform
• Multi-level Feature Fusion-based CNN for Local Climate Zone Classification from Sentinel-2 Images: Benchmark Results on the So2Sat LCZ42 Dataset
• LANCE: Efficient Low-Precision Quantized Winograd Convolution for Neural Networks Based on Graphics Processing Units
• Software-Defined FPGA Accelerator Design for Mobile Deep Learning Applications
• Salient Facial Features from Humans and Deep Neural Networks
• Machine vision detection to daily facial fatigue with a nonlocal 3D attention network
• Progressive Multi-stage Feature Mix for Person Re-Identification
• Explanation of Reinforcement Learning Model in Dynamic Multi-Agent System
• Deep Learning Based Decision Support for Medicine -- A Case Study on Skin Cancer Diagnosis
• Generalization on the Enhancement of Layerwise Relevance Interpretability of Deep Neural Network
• Combining Semantic Guidance and Deep Reinforcement Learning For Generating Human Level Paintings
• Neuron Segmentation Using Deep Complete Bipartite Networks
• Are Saddles Good Enough for Deep Learning?
• A Fully Trainable Network with RNN-based Pooling
• One-Step Time-Dependent Future Video Frame Prediction with a Convolutional Encoder-Decoder Neural Network
• Exploring the potential of transfer learning for metamodels of heterogeneous material deformation
• Slashing Communication Traffic in Federated Learning by Transmitting Clustered Model Updates
• Examining and Mitigating Kernel Saturation in Convolutional Neural Networks using Negative Images
• GAM: Explainable Visual Similarity and Classification via Gradient Activation Maps
• Evaluating the Use of Reconstruction Error for Novelty Localization
• Looking at the posterior: accuracy and uncertainty of neural-network predictions
• Convolution in Convolution for Network in Network
• The Generalizability of Explanations
• Femtosecond pulse parameter estimation from photoelectron momenta using machine learning
• RBUE: A ReLU-Based Uncertainty Estimation Method of Deep Neural Networks
• A backward pass through a CNN using a generative model of its activations
• An Analysis of Human-centered Geolocation
• ForestHash: Semantic Hashing With Shallow Random Forests and Tiny Convolutional Networks
• Unbounded Output Networks for Classification
• Deep Relevance Regularization: Interpretable and Robust Tumor Typing of Imaging Mass Spectrometry Data
• Deep Transform and Metric Learning Networks
• A Resizable Mini-batch Gradient Descent based on a Multi-Armed Bandit
• ConAM: Confidence Attention Module for Convolutional Neural Networks
• AdjointBackMapV2: Precise Reconstruction of Arbitrary CNN Unit's Activation via Adjoint Operators
• Associating Grasp Configurations with Hierarchical Features in Convolutional Neural Networks
• AutoToon: Automatic Geometric Warping for Face Cartoon Generation
• Robust Sampling in Deep Learning
• "You might also like this model": Data Driven Approach for Recommending Deep Learning Models for Unknown Image Datasets
• Semantic Network Interpretation
• Interpreting Attributions and Interactions of Adversarial Attacks
• N2RPP: An Adversarial Network to Rebuild Plantar Pressure for ACLD Patients
• Sorted Pooling in Convolutional Networks for One-shot Learning
• Unsupervised discovery of Interpretable Visual Concepts
• Rethinking Radiology: An Analysis of Different Approaches to BraTS
• ViP: Virtual Pooling for Accelerating CNN-based Image Classification and Object Detection
• In situ TensorView: In situ Visualization of Convolutional Neural Networks
• Implementation of Optical Deep Neural Networks using the Fabry-Perot Interferometer
• Implicit Saliency in Deep Neural Networks
• Saliency strikes back: How filtering out high frequencies improves white-box explanations
• A Framework to Learn with Interpretation
• Learning Propagation Rules for Attribution Map Generation
• Sequentially Aggregated Convolutional Networks
• Diagnostic Visualization for Deep Neural Networks Using Stochastic Gradient Langevin Dynamics
• Interpreting Deep Neural Networks with Relative Sectional Propagation by Analyzing Comparative Gradients and Hostile Activations
• Wider Vision: Enriching Convolutional Neural Networks via Alignment to External Knowledge Bases
• Knowledge Distillation By Sparse Representation Matching
• Mutual Information Preserving Back-propagation: Learn to Invert for Faithful Attribution
• Image Generation and Recognition (Emotions)
• Sequenced-Replacement Sampling for Deep Learning
• End-to-End Race Driving with Deep Reinforcement Learning
• Exploring the influence of scale on artist attribution
• Maximizing Invariant Data Perturbation with Stochastic Optimization
• Learning Navigation by Visual Localization and Trajectory Prediction
• Efficient Modelling Across Time of Human Actions and Interactions
• Interpretation of 3D CNNs for Brain MRI Data Classification