/research

Machine Learning Notes

Machine Learning Notes

Useful Links Of Machine Learning

  • Neutral Network
  • Conference Papers

Neutral Network

Autoencoders

Generative Model

  • GAN: Generative Adversarial Networks (pdf, code)
  • InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Nets (pdf, code)
  • VAE: Auto-Encoding Variational Bayes (pdf, tutorial, tensorflow)
  • PixelRNN: Pixel Recurrent Neural Networks (pdf)
  • PixelCNN: Conditional Image Generation with PixelCNN Decoders (pdf)
  • GMM: Generative Moment Matching Networks (pdf)
  • DARN: Deep AutoRegressive Networks (pdf)

Representation Learning

  • Unsupervised Learning of Object Landmarks through Conditional Image Generation (pdf)

Variants Of Convolution Network

  • Convolution Guide: A guide to convolution arithmetic for deeplearning (pdf)
  • Deformable Convolutional Networks (pdf, mxnet)
  • Spatial Transformer Networks (pdf, tensorflow)

Architeture Of Deep Network

  • VggNet: Very Deep Convolutional Networks for Large-Scale Image Recognition (pdf)
  • ResNet: Deep Residual Learning for Image Recognition (pdf, tensorflow, caffe, mxnet)
  • WRN: Wide residual networks (pdf)
  • ResNext: Aggregated residual transformations for deep neural networks (pdf)
  • Incption
    • v1: Going Deeper with Convolutions (pdf)
    • v2: Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift (pdf)
    • v3: Rethinking the Inception Architecture for Computer Vision (pdf)
    • v4: Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning (pdf)
  • DenseNet: Densely Connected Convolutional Networks (pdf, code)
  • SiameseNet: Siamese Neural Networks for One-shot Image Recognition (pdf)
  • MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications (pdf)
  • MobileNetV2: Inverted Residuals and Linear Bottlenecks (pdf)
  • MobileNetV3
  • Xception: Deep Learning with Depthwise Separable Convolutions (pdf)

Object Detection

  • RCNN: R-CNN: Regions with Convolutional Neural Network Features (pdf)
  • Fast-RCNN: Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks (pdf)
  • Faster-RCNN: Faster R-CNN: Towards Real-Time ObjectDetection with Region Proposal Networks (pdf)
  • R-FCN (pdf)
  • Mask-RCNN (pdf)
  • Focal Loss: Focal Loss for Dense Object Detection (pdf)
  • Non-local Neural Networks(pdf)
  • FPN: Feature Pyramid Networks for Object Detection (pdf)
  • Cascade R-CNN: Delving into High Quality Object Detection (pdf)
  • SNIP: An Analysis of Scale Invariance in Object Detection (pdf)
  • YOLO-3D: Real-Time Seamless Single Shot 6D Object Pose Prediction (pdf)
  • Light-Head R-CNN: In Defense of Two-Stage Object Detector (pdf)
  • FCOS: Fully Convolutional One-Stage Object Detection (pdf)

OCR

  • CRNN: An End-to-End Trainable Neural Network for Image-based Sequence Recognition and Its Application to Scene Text Recognitio (pdf)

Tricks For Training Network

  • Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift (pdf)

Face Recognition

  • DeepID1: Deep Learning Face Representation from Predicting 10,000 Classes (pdf)
  • DeepID2: Deep Learning Face Representation by Joint Identification-Verification (pdf)
  • DeepID2+: Deeply learned face representations are sparse, selective, and robust (pdf)
  • DeepID3: Face recognition with very deep neural networks (pdf)
  • FaceNet: A Unified Embedding for Face Recognition and Clustering (pdf)
  • MTCNN: Joint Face Detection and Alignment using Multi-task Cascaded Convolutional Networks (pdf)

Deep Metric Learning

  • TripletNet: Deep metric learning using Triplet network (pdf)

Uncategorized

  • Deep Photo Style Transfer (pdf, code)
  • Ladder Networks:
    • Lateral Connections in Denoising Autoencoders (pdf)
    • From Neural PCA to Deep Unsupervised Learning (pdf)
    • Semi-Supervised Learning with Ladder Networks (pdf, theano, tensorflow)
    • Deconstructing the Ladder Network Architecture (pdf)
  • Learning Deep Features for Discriminative Localization (pdf, project)
  • A Year In Computer Vision
  • DeepWarp: Photorealistic Image Resynthesis for Gaze Manipulation (pdf, tensorflow)
  • Perceptual Losses for Real-Time Style Transfer (pdf)
  • WESPE: Weakly Supervised Photo Enhancer for Digital Cameras (pdf)

Image Inpainting

  • Context Encoders: Feature Learning by Inpainting (pdf, lua, pytorch)

Transfer Learning

  • CycleGAN: Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks (pdf, office-pytorch)
  • DualGAN: Unsupervised Dual Learning for Image-to-Image Translation (pdf, tensorflow)
  • DiscoGAN: Learning to Discover Cross-Domain Relations with Generative Adversarial Networks (pdf)

Reading List

  • Second Order
    • Is Second-order Information Helpful for Large-scale Visual Recognition (pdf)
    • Second-order Convolutional Neural Networks (pdf)
  • Metric Learning
    • Improved Deep Metric Learning with Multi-class N-pair Loss Objective (pdf)
    • Deep metric learning using Triplet network (pdf)
  • Others
    • DRAW: A Recurrent Neural Network For Image Generation: (pdf)
    • High-Resolution Image Inpainting using Multi-Scale Neural Patch Synthesis (pdf, torch)
    • Reconstruction of Hidden Representation for Robust Feature Extraction (pdf)
    • Between-class Learning for Image Classification (pdf)
    • Age Regression by Conditional Adversarial Autoencoder (pdf, office)
    • Split-Brain Autoencoders: Unsupervised Learning by Cross-Channel Prediction (pdf, office)
    • Stacked Similarity-Aware Autoencoders (pdf)
    • Joint Unsupervised Learning of Deep Representations and Image Clusters (pdf)
    • Lifelong learning with a network of experts (pdf)
    • Crossing Generative Adversarial Networks for Cross-View Person Re-identification (pdf)
    • Deep Unsupervised Clustering Using Mixture of Autoencoders (pdf)
  • Waiting
    • Bilinear CNN Models for Fine-grained Visual Recognition (pdf)
    • Interpretable Transformations with Encoder-Decoder Networks (pdf)
    • Deformable Convolutional Networks (pdf)
    • Learning Hierarchical Features from Generative Models (pdf)
    • Multi-Level Variational Autoencoder: Learning Disentangled Representations from Grouped Observations (pdf)
    • XGAN: Unsupervised Image-to-Image Translation for Many-to-Many Mappings
    • Crossing Generative Adversarial Networks for Cross-View Person Re-identification
    • Deep Unsupervised Clustering Using Mixture of Autoencoders
    • Adversarial Symmetric Variational Autoencoder
    • LapGAN: Deep Generative Image Models using a Laplacian Pyramid of Adversarial Networks (pdf)
  • Recent
    • FSA-Net: Learning Fine-Grained Structure Aggregation for Head Pose Estimation from a Single Image
    • Semantic Alignment: Finding Semantically Consistent Ground-truth for Facial Landmark Detection

Hard Paper

  • Variational Approaches for Auto-Encoding Generative Adversarial Networks (pdf)
  • Nonparametric Inference for Auto-Encoding Variational Bayes (pdf)

Conferentce Paper