- Paper collection about model compression and acceleration:
-
1.Pruning
- 1.1. Filter Pruning
- 1.2. Weight Pruning
-
2.Quantization
- 2.1. Multi-bit Quantization
- 2.2. 1-bit Quantization
-
3.Light-weight Design
-
4.Knowledge Distillation
-
5.Tensor Decomposition
-
6.Other
-
- 2020-CVPRo-HRank: Filter Pruning Using High-Rank Feature Map [Code]
- 2020-CVPR-Learning Filter Pruning Criteria for Deep Convolutional Neural Networks Acceleration
- 2020-CVPR-DMCP: Differentiable Markov Channel Pruning for Neural Networks
- 2020-CVPR-Neural Network Pruning With Residual-Connections and Limited-Data
- 2020-CVPR-GhostNet: More Features from Cheap Operations [Code]
- 2020-CVPR-AdderNet: Do We Really Need Multiplications in Deep Learning? [Code]
- 2020-CVPR-Online Knowledge Distillation via Collaborative Learning
- 2020-CVPR-Regularizing Class-Wise Predictions via Self-Knowledge Distillation
- 2020-CVPR-Explaining Knowledge Distillation by Quantifying the Knowledge
- 2020-CVPR-Dreaming to Distill: Data-Free Knowledge Transfer via DeepInversion [Code]
- 2020-CVPR-Low-rank Compression of Neural Nets: Learning the Rank of Each Layer
- 2020-CVPR-Filter Grafting for Deep Neural Networks
- 2020-CVPR-Structured Compression by Weight Encryption for Unstructured Pruning and Quantization
- 2020-CVPR-APQ: Joint Search for Network Architecture, Pruning and Quantization Policy
- 2020-CVPR-Group Sparsity: The Hinge Between Filter Pruning and Decomposition for Network Compression [Code]
- 2020-CVPR-Multi-Dimensional Pruning: A Unified Framework for Model Compression
- 2020-CVPR-Discrete Model Compression With Resource Constraint for Deep Neural Networks
- 2020-CVPR-Automatic Neural Network Compression by Sparsity-Quantization Joint Learning: A Constrained Optimization-Based Approach
- 2020-CVPR-Low-Rank Compression of Neural Nets: Learning the Rank of Each Layer
- 2020-CVPR-The Knowledge Within: Methods for Data-Free Model Compression
- 2020-CVPR-GAN Compression: Efficient Architectures for Interactive Conditional GANs [Code]
- 2020-CVPR-Few Sample Knowledge Distillation for Efficient Network Compression
- 2020-CVPR-Structured Multi-Hashing for Model Compression
- 2020-CVPRo-Towards Efficient Model Compression via Learned Global Ranking [Code]
- 2020-CVPR-Training Quantized Neural Networks With a Full-Precision Auxiliary Module
- 2020-CVPR-Adaptive Loss-aware Quantization for Multi-bit Networks
- 2020-CVPR-ZeroQ: A Novel Zero Shot Quantization Framework
- 2020-CVPR-BiDet: An Efficient Binarized Object Detector [code]
- 2020-CVPR-Forward and Backward Information Retention for Accurate Binary Neural Networks
- 2020-CVPR-Binarizing MobileNet via Evolution-Based Searching
- 2020-CVPR-Collaborative Distillation for Ultra-Resolution Universal Style Transfer [Code]
- 2020-CVPR-Self-training with Noisy Student improves ImageNet classification [Code]
- 2020-CVPR-Neural Networks Are More Productive Teachers Than Human Raters: Active Mixup for Data-Efficient Knowledge Distillation From a Blackbox Model
- 2020-CVPR-Heterogeneous Knowledge Distillation Using Information Flow Modeling
- 2020-CVPR-Revisiting Knowledge Distillation via Label Smoothing Regularization
- 2020-CVPR-Distilling Knowledge From Graph Convolutional Networks
- 2020-CVPR-MineGAN: Effective Knowledge Transfer From GANs to Target Domains With Few Images [Code]
- 2020-CVPR-Distilling Cross-Task Knowledge via Relationship Matching
- 2020-ECCV-EagleEye: Fast Sub-net Evaluation for Efficient Neural Network Pruning
- 2020-ECCV-ReActNet: Towards Precise Binary Neural Network with Generalized Activation Functions
- 2020-ECCV-Knowledge Distillation Meets Self-Supervision
- 2020-ECCV-Differentiable Feature Aggregation Search for Knowledge Distillation
- 2020-ECCV-Post-Training Piecewise Linear Quantization for Deep Neural Networks
- 2020-ECCV-GAN Slimming: All-in-One GAN Compression by A Unified Optimization Framework
- 2020-ECCV-Online Ensemble Model Compression using Knowledge Distillation
- 2020-ECCV-Stable Low-rank Tensor Decomposition for Compression of Convolutional Neural Network
- 2020-ECCV-DSA: More Efficient Budgeted Pruning via Differentiable Sparsity Allocation
- 2020-ECCV-Accelerating CNN Training by Pruning Activation Gradients
- 2020-ECCV-DHP: Differentiable Meta Pruning via HyperNetworks
- 2020-ECCV-Differentiable Joint Pruning and Quantization for Hardware Efficiency
- 2020-ECCV-Meta-Learning with Network Pruning
- 2020-ECCV-BATS: Binary ArchitecTure Search
- 2020-ECCV-Learning Architectures for Binary Networks
- 2020-ECCV-DA-NAS: Data Adapted Pruning for Efficient Neural Architecture Search
- 2020-ECCV-Knowledge Transfer via Dense Cross-Layer Mutual-Distillation
- 2020-ECCV-Generative Low-bitwidth Data Free Quantization
- 2020-ECCV-HMQ: Hardware Friendly Mixed Precision Quantization Block for CNNs
- 2020-ECCV-Search What You Want: Barrier Panelty NAS for Mixed Precision Quantization
- 2020-ECCV-Rethinking Bottleneck Structure for Efficient Mobile Network Design
- 2020-ECCV-PSConv: Squeezing Feature Pyramid into One Compact Poly-Scale Convolutional Layer
- ...
- 2020-NIPS-Rotated Binary Neural Network [code]
- 2020-NIPS-Path Sample-Analytic Gradient Estimators for Stochastic Binary Networks
- 2020-NIPS-Efficient Exact Verification of Binarized Neural Networks
- 2020-NIPS-Reintroducing Straight-Through Estimators asPrincipled Methods for Stochastic Binary Networks
- 2020-NIPS-Sanity-Checking Pruning Methods: Random Tickets can Win the Jackpot
- 2020-NIPS-Network Pruning via Greedy Optimization: Fast Rate and Efficient Algorithm
- 2020-NIPS-Neuron Merging: Compensating for Pruned Neurons
- 2020-NIPS-Scientific Control for Reliable Neural Network Pruning
- 2020-NIPS-Neuron-level Structured Pruning using Polarization Regularizer
- 2020-NIPS-Directional Pruning of Deep Neural Networks
- 2020-NIPS-Storage Efficient and Dynamic Flexible Runtime Channel Pruning via Deep Reinforcement Learning
- 2020-NIPS-Movement Pruning: Adaptive Sparsity by Fine-Tuning
- 2020-NIPS-The Generalization-Stability Tradeoff In Neural Network Pruning
- 2020-NIPS-Pruning neural networks without any data by conserving synaptic flow
- 2020-NIPS-Network Pruning via Greedy Optimization: Fast Rate and Efficient Algorithms
- 2020-NIPS-HYDRA: Pruning Adversarially Robust Neural Networks
- 2020-NIPS-Logarithmic Pruning is All You Need
- 2020-NIPS-Pruning Filter in Filter
- 2020-NIPS-Bayesian Bits: Unifying Quantization and Pruning
- 2020-NIPS-Searching for Low-Bit Weights in Quantized Neural Networks
- 2020-NIPS-Robust Quantization: One Model to Rule Them All
- 2020-NIPS-Position-based Scaled Gradient for Model Quantization and Sparse Training
- 2020-NIPS-Universally Quantized Neural Compression
- 2020-NIPS-FleXOR: Trainable Fractional Quantization
- 2020-NIPS-Efficient Exact Verification of Binarized Neural Networks
- 2020-NIPS-Path Sample-Analytic Gradient Estimators for Stochastic Binary Networks
- 2020-NIPS-WoodFisher: Efficient Second-Order Approximation for Neural Network Compression
- 2020-NIPS-Knowledge Distillation in Wide Neural Networks: Risk Bound, Data Efficiency and Imperfect Teacher
- 2020-NIPS-Task-Oriented Feature Distillation
- 2020-NIPS-Cream of the Crop: Distilling Prioritized Paths For One-Shot Neural Architecture Search
- 2020-NIPS-Self-Distillation Amplifies Regularization in Hilbert Space
- 2020-NIPS-MiniLM: Deep Self-Attention Distillation for Task-Agnostic Compression of Pre-Trained Transformers
- 2020-NIPS-Self-Distillation as Instance-Specific Label Smoothing
- 2020-NIPS-Agree to Disagree: Adaptive Ensemble Knowledge Distillation in Gradient Space
- 2020-NIPS-Distributed Distillation for On-Device Learning
- 2020-NIPS-Residual Distillation: Towards Portable Deep Neural Networks without Shortcuts
- 2020-NIPS-Ensemble Distillation for Robust Model Fusion in Federated Learning
- 2020-NIPS-Fast, Accurate, and Simple Models for Tabular Data via Augmented Distillation
- 2020-ICML-PENNI: Pruned Kernel Sharing for Efficient CNN Inference
- 2020-ICML-Operation-Aware Soft Channel Pruning using Differentiable Masks
- 2020-ICML-DropNet: Reducing Neural Network Complexity via Iterative Pruning
- 2020-ICML-Proving the Lottery Ticket Hypothesis: Pruning is All You Need
- 2020-ICML-Network Pruning by Greedy Subnetwork Selection
- 2020-ICML-AutoGAN-Distiller: Searching to Compress Generative Adversarial Networks
- 2020-ICML-Adversarial Neural Pruning with Latent Vulnerability Suppression
- 2020-ICML-Feature-map-level Online Adversarial Knowledge Distillation
- 2020-ICML-Knowledge transfer with jacobian matching
- 2020-ICML-Good Subnetworks Provably Exist Pruning via Greedy Forward Selection
- 2020-ICML-Training Binary Neural Networks through Learning with Noisy Supervision
- 2020-ICML-Multi-Precision Policy Enforced Training (MuPPET) : A Precision-Switching Strategy for Quantised Fixed-Point Training of CNNs
- 2020-ICML-Divide and Conquer: Leveraging Intermediate Feature Representations for Quantized Training of Neural Networks
- 2020-ICML-Feature Quantization Improves GAN Training [code]
- 2020-ICML-Towards Accurate Post-training Network Quantization via Bit-Split and Stitching
- 2020-ICML-Accelerating Large-Scale Inference with Anisotropic Vector Quantization
- 2020-ICML-Differentiable Product Quantization for Learning Compact Embedding Layers [code]
- 2020-ICML-Up or Down? Adaptive Rounding for Post-Training Quantization
- 2020-ICLR-Lookahead: A Far-sighted Alternative of Magnitude-based Pruning [Code]
- 2020-ICLR-Dynamic Model Pruning with Feedback
- 2020-ICLR-Provable Filter Pruning for Efficient Neural Networks
- 2020-ICLR-Data-Independent Neural Pruning via Coresets
- 2020-ICLR-FSNet: Compression of Deep Convolutional Neural Networks by Filter Summary
- 2020-ICLR-Probabilistic Connection Importance Inference and Lossless Compression of Deep Neural Networks
- 2020-ICLR-BlockSwap: Fisher-guided Block Substitution for Network Compression on a Budget
- 2020-ICLR-Neural Epitome Search for Architecture-Agnostic Network Compression
- 2020-ICLR-One-Shot Pruning of Recurrent Neural Networks by Jacobian Spectrum Evaluation
- 2020-ICLR-DeepHoyer: Learning Sparser Neural Network with Differentiable Scale-Invariant Sparsity Measures [Code]
- 2020-ICLR-Dynamic Sparse Training: Find Efficient Sparse Network From Scratch With Trainable Masked Layers
- 2020-ICLR-Scalable Model Compression by Entropy Penalized Reparameterization
- 2020-ICLR-One-shot Pruning of Recurrent Neural Neworks by Jacobian Spectrum Evaluation
- 2020-ICLR-Mixed Precision DNNs: All you need is a good parametrization
- 2020-ICLR-Comparing Fine-tuning and Rewinding in Neural Network Pruning
- 2020-ICLR-A Signal Propagation Perspective for Pruning Neural Networks at Initialization
- 2020-ICLR-Linear Symmetric Quantization of Neural Networks for Low-precision Integer Hardware
- 2020-ICLR-AutoQ: Automated Kernel-Wise Neural Network Quantization
- 2020-ICLR-Additive Powers-of-Two Quantization: A Non-uniform Discretization for Neural Networks
- 2020-ICLR-Learned Step Size Quantization
- 2020-ICLR-Sampling-Free Learning of Bayesian Quantized Neural Networks
- 2020-ICLR-Gradient $\ell_1$ Regularization for Quantization Robustness
- 2020-ICLR-BinaryDuo: Reducing Gradient Mismatch in Binary Activation Network by Coupling Binary Activations [code]
- 2020-ICLR-Training binary neural networks with real-to-binary convolutions [code]
- 2020-ICLR-Critical initialisation in continuous approximations of binary neural networks
- 2020-ICLR-Comparing Rewinding and Fine-tuning in Neural Network Pruning
- 2020-ICLR-ProxSGD: Training Structured Neural Networks under Regularization and Constraints
- ...
- 2020-AAAI-Binarized Neural Architecture Search
- 2019-CVPRo-HAQ: hardware-aware automated quantization
- 2019-CVPRo-Filter Pruning via Geometric Median for Deep Convolutional Neural Networks Acceleration [Code]
- 2019-CVPR-All You Need is a Few Shifts: Designing Efficient Convolutional Neural Networks for Image Classification
- 2019-CVPR-Importance Estimation for Neural Network Pruning [Code]
- 2019-CVPR-HetConv Heterogeneous Kernel-Based Convolutions for Deep CNNs
- 2019-CVPR-Fully Learnable Group Convolution for Acceleration of Deep Neural Networks
- 2019-CVPR-Towards Optimal Structured CNN Pruning via Generative Adversarial Learning
- 2019-CVPR-ChamNet: Towards Efficient Network Design through Platform-Aware Model Adaptation
- 2019-CVPR-Partial Order Pruning: for Best Speed/Accuracy Trade-off in Neural Architecture Search [Code]
- 2019-CVPR-Auto-DeepLab: Hierarchical Neural Architecture Search for Semantic Image Segmentation [Code]
- 2019-CVPR-MnasNet: Platform-Aware Neural Architecture Search for Mobile [Code]
- 2019-CVPR-MFAS: Multimodal Fusion Architecture Search
- 2019-CVPR-A Neurobiological Evaluation Metric for Neural Network Model Search
- 2019-CVPR-Fast Neural Architecture Search of Compact Semantic Segmentation Models via Auxiliary Cells
- 2019-CVPR-Efficient Neural Network Compression [Code]
- 2019-CVPR-T-Net: Parametrizing Fully Convolutional Nets with a Single High-Order Tensor
- 2019-CVPR-Centripetal SGD for Pruning Very Deep Convolutional Networks with Complicated Structure [Code]
- 2019-CVPR-DSC: Dense-Sparse Convolution for Vectorized Inference of Convolutional Neural Networks
- 2019-CVPR-DupNet: Towards Very Tiny Quantized CNN With Improved Accuracy for Face Detection
- 2019-CVPR-ECC: Platform-Independent Energy-Constrained Deep Neural Network Compression via a Bilinear Regression Model
- 2019-CVPR-Variational Convolutional Neural Network Pruning
- 2019-CVPR-Accelerating Convolutional Neural Networks via Activation Map Compression
- 2019-CVPR-Compressing Convolutional Neural Networks via Factorized Convolutional Filters
- 2019-CVPR-Deep Virtual Networks for Memory Efficient Inference of Multiple Tasks
- 2019-CVPR-Exploiting Kernel Sparsity and Entropy for Interpretable CNN Compression
- 2019-CVPR-MBS: Macroblock Scaling for CNN Model Reduction
- 2019-CVPR-On Implicit Filter Level Sparsity in Convolutional Neural Networks
- 2019-CVPR-Structured Pruning of Neural Networks With Budget-Aware Regularization
- 2019-CVPRo-Neural Rejuvenation: Improving Deep Network Training by Enhancing Computational Resource Utilization [Code]
- 2019-CVPR-Knowledge Representing: Efficient, Sparse Representation of Prior Knowledge for Knowledge Distillation
- 2019-CVPR-Knowledge Distillation via Instance Relationship Graph
- 2019-CVPR-Variational Information Distillation for Knowledge Transfer
- 2019-CVPR-Learning Metrics from Teachers Compact Networks for Image Embedding [Code]
- 2019-CVPR-Enhanced Bayesian Compression via Deep Reinforcement Learning
- 2019-CVPR-Cross Domain Model Compression by Structurally Weight Sharing
- 2019-CVPR-Cascaded Projection: End-To-End Network Compression and Acceleration
- 2019-CVPR-Fully Quantized Network for Object Detection
- 2019-CVPR-Learning to Quantize Deep Networks by Optimizing Quantization Intervals With Task Loss
- 2019-CVPR-Quantization Networks
- 2019-CVPR-SeerNet: Predicting Convolutional Neural Network Feature-Map Sparsity Through Low-Bit Quantization
- 2019-CVPR-Simultaneously Optimizing Weight and Quantizer of Ternary Neural Network Using Truncated Gaussian Approximation
- 2019-CVPR-Binary Ensemble Neural Network: More Bits per Network or More Networks per Bit?
- 2019-CVPR-A Main/Subsidiary Network Framework for Simplifying Binary Neural Networks
- 2019-CVPR-Regularizing Activation Distribution for Training Binarized Deep Networks
- 2019-CVPR-Structured Binary Neural Networks for Accurate Image Classification and Semantic Segmentation
- 2019-CVPR-Learning Channel-Wise Interactions for Binary Convolutional Neural Networks
- 2019-CVPR-Circulant Binary Convolutional Networks: Enhancing the Performance of 1-Bit DCNNs With Circulant Back Propagation
- 2019-CVPR-HAQ: Hardware-Aware Automated Quantization With Mixed Precision
- 2019-CVPR-ECC: Platform-Independent Energy-Constrained Deep Neural Network Compression via a Bilinear Regression Model
- 2019-CVPR-OICSR: Out-In-Channel Sparsity Regularization for Compact Deep Neural Networks
- ...
- 2019-ICCV-Rethinking ImageNet Pre-training
- 2019-ICCV-Universally Slimmable Networks and Improved Training Techniques
- 2019-ICCV-MetaPruning: Meta Learning for Automatic Neural Network Channel Pruning [Code]
- 2019-ICCV-Progressive Differentiable Architecture Search: Bridging the Depth Gap between Search and Evaluation [Code]
- 2019-ICCV-ACNet: Strengthening the Kernel Skeletons for Powerful CNN via Asymmetric Convolution Blocks
- 2019-ICCV-A Comprehensive Overhaul of Feature Distillation
- 2019-ICCV-Similarity-Preserving Knowledge Distillation
- 2019-ICCV-Correlation Congruence for Knowledge Distillation
- 2019-ICCV-Data-Free Learning of Student Networks
- 2019-ICCV-Learning Lightweight Lane Detection CNNs by Self Attention Distillation [Code]
- 2019-ICCV-Attention bridging network for knowledge transfer
- 2019-ICCV-Learning Filter Basis for Convolutional Neural Network Compression
- 2019-ICCV-Accelerate CNN via Recursive Bayesian Pruning
- 2019-ICCV-Adversarial Robustness vs Model Compression, or Both?
- ...
- 2019-NIPS-Global Sparse Momentum SGD for Pruning Very Deep Neural Networks
- 2019-NIPS-Model Compression with Adversarial Robustness: A Unified Optimization Framework
- 2019-NIPS-AutoPrune: Automatic Network Pruning by Regularizing Auxiliary Parameters
- 2019-NIPS-Double Quantization for Communication-Efficient Distributed Optimization
- 2019-NIPS-Focused Quantization for Sparse CNNs
- 2019-NIPS-E2-Train: Training State-of-the-art CNNs with Over 80% Energy Savings
- 2019-NIPS-MetaQuant: Learning to Quantize by Learning to Penetrate Non-differentiable Quantization
- 2019-NIPS-Random Projections with Asymmetric Quantization
- 2019-NIPS-Network Pruning via Transformable Architecture Search [Code]
- 2019-NIPS-Point-Voxel CNN for Efficient 3D Deep Learning [Code]
- 2019-NIPS-Gate Decorator: Global Filter Pruning Method for Accelerating Deep Convolutional Neural Networks
- 2019-NIPS-A Mean Field Theory of Quantized Deep Networks: The Quantization-Depth Trade-Off
- 2019-NIPS-Qsparse-local-SGD: Distributed SGD with Quantization, Sparsification and Local Computations
- 2019-NIPS-Post training 4-bit quantization of convolutional networks for rapid-deployment
- 2019-NIPS-Zero-shot Knowledge Transfer via Adversarial Belief Matching [Code] (spotlight)
- 2019-NIPS-Efficient and Effective Quantization for Sparse DNNs
- 2019-NIPS-Latent Weights Do Not Exist: Rethinking Binarized Neural Network Optimization [paper]
- 2019-NIPS-Post-training 4-bit quantization of convolution networks for rapid-deployment
- 2019-NIPS-PowerSGD: Practical Low-Rank Gradient Compression for Distributed Optimization [paper]
- 2019-NIPS-Deconstructing Lottery Tickets: Zeros, Signs, and the Supermask
- 2019-NIPS-Channel Gating Neural Network
- 2019-NIPS-Positive-Unlabeled Compression on the Cloud [paper]
- 2019-NIPS-Einconv: Exploring Unexplored Tensor Decompositions for Convolutional Neural Networks [paper] [codes]
- 2019-NIPS-A Tensorized Transformer for Language Modeling [paper]
- 2019-NIPS-Shallow RNN: Accurate Time-series Classification on Resource Constrained Devices [paper]
- 2019-NIPS-CondConv: Conditionally Parameterized Convolutions for Efficient Inference [paper]
- 2019-NIPS-SCAN: A Scalable Neural Networks Framework Towards Compact and Efficient Models [paper]
- 2019-NIPS-Hybrid 8-bit Floating Point (HFP8) Training and Inference for Deep Neural Networks [paper]
- 2019-NIPS-Backprop with Approximate Activations for Memory-efficient Network Training [paper]
- 2019-NIPS-Dimension-Free Bounds for Low-Precision Training
- 2019-NIPS-One ticket to win them all: generalizing lottery ticket initializations across datasets and optimizers
- ...
- 2019-ICML-Approximated Oracle Filter Pruning for Destructive CNN Width Optimization [Code]
- 2019-ICML-EigenDamage: Structured Pruning in the Kronecker-Factored Eigenbasis [Code]
- 2019-ICML-Zero-Shot Knowledge Distillation in Deep Networks [Code]
- 2019-ICML-LegoNet: Efficient Convolutional Neural Networks with Lego Filters [Code]
- 2019-ICML-EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks [Code]
- 2019-ICML-Collaborative Channel Pruning for Deep Networks
- 2019-ICML-Training CNNs with Selective Allocation of Channels
- 2019-ICML-NAS-Bench-101: Towards Reproducible Neural Architecture Search [Code]
- 2019-ICMLw-Towards Learning of Filter-Level Heterogeneous Compression of Convolutional Neural Networks [Code] (AutoML workshop)
- 2019-ICML-Improving Neural Network Quantization without Retraining using Outlier Channel Splitting
- 2019-ICML-Same, Same But Different: Recovering Neural Network Quantization Error Through Weight Factorization
- 2019-ICML-Parameter efficient training of deep convolutional neural networks by dynamic sparse reparameterization
- 2019-ICML-LIT: Learned Intermediate Representation Training for Model Compression
- 2019-ICML-Towards Understanding Knowledge Distillation
- 2019-ICML-Rate Distortion For Model Compression: From Theory To Practice
- 2019-ICML-Approximated Oracle Filter Pruning for Destructive CNN Width Optimization github
- ...
- 2019-ICLRo-The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networks (best paper!)
- 2019-ICLR-Slimmable Neural Networks [Code]
- 2019-ICLR-Defensive Quantization: When Efficiency Meets Robustness
- 2019-ICLR-Minimal Random Code Learning: Getting Bits Back from Compressed Model Parameters [Code]
- 2019-ICLR-ProxylessNAS: Direct Neural Architecture Search on Target Task and Hardware [Code]
- 2019-ICLR-SNIP: Single-shot Network Pruning based on Connection Sensitivity
- 2019-ICLR-Non-vacuous Generalization Bounds at the ImageNet Scale: a PAC-Bayesian Compression Approach
- 2019-ICLR-Dynamic Channel Pruning: Feature Boosting and Suppression
- 2019-ICLR-Energy-Constrained Compression for Deep Neural Networks via Weighted Sparse Projection and Layer Input Masking
- 2019-ICLR-RotDCF: Decomposition of Convolutional Filters for Rotation-Equivariant Deep Networks
- 2019-ICLR-Dynamic Sparse Graph for Efficient Deep Learning
- 2019-ICLR-Big-Little Net: An Efficient Multi-Scale Feature Representation for Visual and Speech Recognition
- 2019-ICLR-Data-Dependent Coresets for Compressing Neural Networks with Applications to Generalization Bounds
- 2019-ICLR-Learning Recurrent Binary/Ternary Weights
- 2019-ICLR-Double Viterbi: Weight Encoding for High Compression Ratio and Fast On-Chip Reconstruction for Deep Neural Network
- 2019-ICLR-Relaxed Quantization for Discretized Neural Networks
- 2019-ICLR-Integer Networks for Data Compression with Latent-Variable Models
- 2019-ICLR-Minimal Random Code Learning: Getting Bits Back from Compressed Model Parameters
- 2019-ICLR-Analysis of Quantized Models
- 2019-ICLR-DARTS: Differentiable Architecture Search [Code]
- 2019-ICLR-Graph HyperNetworks for Neural Architecture Search
- 2019-ICLR-Learnable Embedding Space for Efficient Neural Architecture Compression [Code]
- 2019-ICLR-Efficient Multi-Objective Neural Architecture Search via Lamarckian Evolution
- 2019-ICLR-SNAS: stochastic neural architecture search
- 2019-ICLR-Integral Pruning on Activations and Weights for Efficient Neural Networks
- 2019-ICLR-Rethinking the Value of Network Pruning
- 2019-ICLR-ProxQuant: Quantized Neural Networks via Proximal Operators
- 2019-ICLR-Per-Tensor Fixed-Point Quantization of the Back-Propagation Algorithm
- 2019-ICLR-Combinatorial Attacks on Binarized Neural Networks
- 2019-ICLR-ARM: Augment-REINFORCE-Merge Gradient for Stochastic Binary Networks
- 2019-ICLR-An Empirical study of Binary Neural Networks' Optimisation
- 2019-ICLR-On the Universal Approximability and Complexity Bounds of Quantized ReLU Neural Networks
- 2019-ICLR-Understanding Straight-Through Estimator in Training Activation Quantized Neural Nets
- ...
- 2018-CVPR-Context-Aware Deep Feature Compression for High-Speed Visual Tracking
- 2018-CVPR-NISP: Pruning Networks using Neuron Importance Score Propagation
- 2018-CVPR-Condensenet: An efficient densenet using learned group convolutions [Code]
- 2018-CVPR-Shift: A zero flop, zero parameter alternative to spatial convolutions
- 2018-CVPR-Explicit Loss-Error-Aware Quantization for Low-Bit Deep Neural Networks
- 2018-CVPR-Interleaved structured sparse convolutional neural networks
- 2018-CVPR-Towards Effective Low-bitwidth Convolutional Neural Networks
- 2018-CVPR-CLIP-Q: Deep Network Compression Learning by In-Parallel Pruning-Quantization
- 2018-CVPR-Blockdrop: Dynamic inference paths in residual networks
- 2018-CVPR-Nestednet: Learning nested sparse structures in deep neural networks
- 2018-CVPR-Stochastic downsampling for cost-adjustable inference and improved regularization in convolutional networks
- 2018-CVPR-Wide Compression: Tensor Ring Nets
- 2018-CVPR-Learning Compact Recurrent Neural Networks With Block-Term Tensor Decomposition
- 2018-CVPR-Learning Time/Memory-Efficient Deep Architectures With Budgeted Super Networks
- 2018-CVPR-HydraNets: Specialized Dynamic Architectures for Efficient Inference
- 2018-CVPR-SYQ: Learning Symmetric Quantization for Efficient Deep Neural Networks
- 2018-CVPR-Towards Effective Low-Bitwidth Convolutional Neural Networks
- 2018-CVPR-Two-Step Quantization for Low-Bit Neural Networks
- 2018-CVPR-Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inference
- 2018-CVPR-"Learning-Compression" Algorithms for Neural Net Pruning
- 2018-CVPR-PackNet: Adding Multiple Tasks to a Single Network by Iterative Pruning [Code]
- 2018-CVPR-MorphNet: Fast & Simple Resource-Constrained Structure Learning of Deep Networks [Code]
- 2018-CVPR-ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices
- 2018-CVPRw-Squeezenext: Hardware-aware neural network design
- 2018-CVPR-Quantization of Fully Convolutional Networks for Accurate Biomedical Image Segmentation
- ...
- 2018-ECCV-A Systematic DNN Weight Pruning Framework using Alternating Direction Method of Multipliers
- 2018-ECCV-Coreset-Based Neural Network Compression
- 2018-ECCV-Data-Driven Sparse Structure Selection for Deep Neural Networks [Code]
- 2018-ECCV-Training Binary Weight Networks via Semi-Binary Decomposition
- 2018-ECCV-Learning Compression from Limited Unlabeled Data
- 2018-ECCV-Constraint-Aware Deep Neural Network Compression
- 2018-ECCV-Sparsely Aggregated Convolutional Networks
- 2018-ECCV-Deep Expander Networks: Efficient Deep Networks from Graph Theory [Code]
- 2018-ECCV-SparseNet-Sparsely Aggregated Convolutional Networks [Code]
- 2018-ECCV-Ask, acquire, and attack: Data-free uap generation using class impressions
- 2018-ECCV-Netadapt: Platform-aware neural network adaptation for mobile applications
- 2018-ECCV-Clustering Convolutional Kernels to Compress Deep Neural Networks
- 2018-ECCV-Bi-Real Net: Enhancing the Performance of 1-bit CNNs With Improved Representational Capability and Advanced Training Algorithm
- 2018-ECCV-Extreme Network Compression via Filter Group Approximation
- 2018-ECCV-Convolutional Networks with Adaptive Inference Graphs
- 2018-ECCV-SkipNet: Learning Dynamic Routing in Convolutional Networks [Code]
- 2018-ECCV-Value-aware Quantization for Training and Inference of Neural Networks
- 2018-ECCV-LQ-Nets: Learned Quantization for Highly Accurate and Compact Deep Neural Networks
- 2018-ECCV-AMC: AutoML for Model Compression and Acceleration on Mobile Devices
- 2018-ECCV-Piggyback: Adapting a single network to multiple tasks by learning to mask weights
- ...
- 2018-NIPS-Discrimination-aware Channel Pruning for Deep Neural Networks
- 2018-NIPS-Frequency-Domain Dynamic Pruning for Convolutional Neural Networks
- 2018-NIPS-ChannelNets: Compact and Efficient Convolutional Neural Networks via Channel-Wise Convolutions
- 2018-NIPS-DropBlock: A regularization method for convolutional networks
- 2018-NIPS-Constructing fast network through deconstruction of convolution
- 2018-NIPS-Learning Versatile Filters for Efficient Convolutional Neural Networks [Code]
- 2018-NIPS-Moonshine: Distilling with cheap convolutions
- 2018-NIPS-HitNet: hybrid ternary recurrent neural network
- 2018-NIPS-FastGRNN: A Fast, Accurate, Stable and Tiny Kilobyte Sized Gated Recurrent Neural Network
- 2018-NIPS-Training DNNs with Hybrid Block Floating Point
- 2018-NIPS-Reversible Recurrent Neural Networks
- 2018-NIPS-Synaptic Strength For Convolutional Neural Network
- 2018-NIPS-Learning sparse neural networks via sensitivity-driven regularization
- 2018-NIPS-Multi-Task Zipping via Layer-wise Neuron Sharing
- 2018-NIPS-A Linear Speedup Analysis of Distributed Deep Learning with Sparse and Quantized Communication
- 2018-NIPS-Gradient Sparsification for Communication-Efficient Distributed Optimization
- 2018-NIPS-GradiVeQ: Vector Quantization for Bandwidth-Efficient Gradient Aggregation in Distributed CNN Training
- 2018-NIPS-ATOMO: Communication-efficient Learning via Atomic Sparsification
- 2018-NIPS-Norm matters: efficient and accurate normalization schemes in deep networks
- 2018-NIPS-Sparsified SGD with memory
- 2018-NIPS-Pelee: A Real-Time Object Detection System on Mobile Devices
- 2018-NIPS-Scalable methods for 8-bit training of neural networks
- 2018-NIPS-TETRIS: TilE-matching the TRemendous Irregular Sparsity
- 2018-NIPS-Training deep neural networks with 8-bit floating point numbers
- 2018-NIPS-Multiple instance learning for efficient sequential data classification on resource-constrained devices
- 2018-NIPSw-Pruning neural networks: is it time to nip it in the bud?
- 2018-NIPSwb-Rethinking the Value of Network Pruning [2019 ICLR version]
- 2018-NIPSw-Structured Pruning for Efficient ConvNets via Incremental Regularization [2019 IJCNN version] [Code]
- 2018-NIPSw-Adaptive Mixture of Low-Rank Factorizations for Compact Neural Modeling
- 2018-NIPSw-Learning Sparse Networks Using Targeted Dropout [OpenReview] [Code]
- 2018-NIPS-BinGAN: Learning Compact Binary Descriptors with a Regularized GAN
- 2018-NIPS-Paraphrasing Complex Network: Network Compression via Factor Transfer
- ...
- 2018-ICML-Compressing Neural Networks using the Variational Information Bottleneck
- 2018-ICML-DCFNet: Deep Neural Network with Decomposed Convolutional Filters
- 2018-ICML-Deep k-Means Re-Training and Parameter Sharing with Harder Cluster Assignments for Compressing Deep Convolutions
- 2018-ICML-Error Compensated Quantized SGD and its Applications to Large-scale Distributed Optimization
- 2018-ICML-High Performance Zero-Memory Overhead Direct Convolutions
- 2018-ICML-Kronecker Recurrent Units
- 2018-ICML-Weightless: Lossy weight encoding for deep neural network compression
- 2018-ICML-StrassenNets: Deep learning with a multiplication budget
- 2018-ICML-Learning Compact Neural Networks with Regularization
- 2018-ICML-WSNet: Compact and Efficient Networks Through Weight Sampling
- 2018-ICML-Gradually Updated Neural Networks for Large-Scale Image Recognition [Code]
- ...
- 2018-ICLRo-Training and Inference with Integers in Deep Neural Networks
- 2018-ICLR-Rethinking the Smaller-Norm-Less-Informative Assumption in Channel Pruning of Convolution Layers
- 2018-ICLR-N2N learning: Network to Network Compression via Policy Gradient Reinforcement Learning
- 2018-ICLR-Model compression via distillation and quantization
- 2018-ICLR-Towards Image Understanding from Deep Compression Without Decoding
- 2018-ICLR-Deep Gradient Compression: Reducing the Communication Bandwidth for Distributed Training
- 2018-ICLR-Mixed Precision Training of Convolutional Neural Networks using Integer Operations
- 2018-ICLR-Mixed Precision Training
- 2018-ICLR-Apprentice: Using Knowledge Distillation Techniques To Improve Low-Precision Network Accuracy
- 2018-ICLR-Loss-aware Weight Quantization of Deep Networks
- 2018-ICLR-Alternating Multi-bit Quantization for Recurrent Neural Networks
- 2018-ICLR-Adaptive Quantization of Neural Networks
- 2018-ICLR-Variational Network Quantization
- 2018-ICLR-Espresso: Efficient Forward Propagation for Binary Deep Neural Networks
- 2018-ICLR-Learning to share: Simultaneous parameter tying and sparsification in deep learning
- 2018-ICLR-Learning Sparse Neural Networks through L0 Regularization
- 2018-ICLR-WRPN: Wide Reduced-Precision Networks
- 2018-ICLR-Deep rewiring: Training very sparse deep networks
- 2018-ICLR-Efficient sparse-winograd convolutional neural networks [Code]
- 2018-ICLR-Learning Intrinsic Sparse Structures within Long Short-term Memory
- 2018-ICLR-Multi-scale dense networks for resource efficient image classification
- 2018-ICLR-Compressing Word Embedding via Deep Compositional Code Learning
- 2018-ICLR-Learning Discrete Weights Using the Local Reparameterization Trick
- 2018-ICLR-Training wide residual networks for deployment using a single bit for each weight
- 2018-ICLR-The High-Dimensional Geometry of Binary Neural Networks
- 2018-ICLRw-To Prune, or Not to Prune: Exploring the Efficacy of Pruning for Model Compression (Similar topic: 2018-NIPSw-nip in the bud, 2018-NIPSw-rethink)
- 2018-ICLRw-Systematic Weight Pruning of DNNs using Alternating Direction Method of Multipliers
- 2018-ICLRw-Weightless: Lossy weight encoding for deep neural network compression
- 2018-ICLRw-Variance-based Gradient Compression for Efficient Distributed Deep Learning
- 2018-ICLRw-Stacked Filters Stationary Flow For Hardware-Oriented Acceleration Of Deep Convolutional Neural Networks
- 2018-ICLRw-Training Shallow and Thin Networks for Acceleration via Knowledge Distillation with Conditional Adversarial Networks
- 2018-ICLRw-Accelerating Neural Architecture Search using Performance Prediction
- 2018-ICLRw-Nonlinear Acceleration of CNNs
- 2018-ICLRw-Attention-Based Guided Structured Sparsity of Deep Neural Networks [Code]
- 2018-ICLR-Stochastic activation pruning for robust adversarial defense
- ...
https://github.com/MingSun-Tse/EfficientDNNs
https://github.com/danielmcpark/awesome-pruning-acceleration
https://github.com/csyhhu/Awesome-Deep-Neural-Network-Compression