This is a curated list of resources that I have found useful regarding machine learning, in particular deep learning. Only books that add significant value to understanding the topic are listed. Also, a list of good articles and some other resources.
-
Introduction to Mathematical Statistics
Robert Hogg, Joeseph McKean and Allen Craig -
Numerical Linear Algebra and Applications
Biswa Nath Datta
- Introduction To Machine Learning
Ethem Alpaydin
-
Reinforcement Learning: An Introduction
Richard Sutton and Andrew Barto -
The Elements of Statistical Learning
Trevor Hastie, Robert Tibshirani and Jerome Friedman -
Pattern Recognition and Machine Learning
Christopher Bishop -
Machine Learning: A Probabilistic Perspective
Kevin Murphy
-
Deep Learning
Ian Goodfellow, Yoshua Bengio and Aaron Courville -
Deep Learning with Python
François Chollet
-
Distributional Multivariate Policy Evaluation and Exploration with the Bellman GAN
-
TensorFuzz: Debugging Neural Networks with Coverage-Guided Fuzzing
-
The challenge of realistic music generation: modelling raw audio at scale
-
Understanding and Improving Interpolation in Autoencoders via an Adversarial Regularizer
-
Reliable Uncertainty Estimates in Deep Neural Networks using Noise Contrastive Priors
-
The Variational Homoencoder: Learning to learn high capacity generative models from few examples
-
Variational Bayesian Reinforcement Learning with Regret Bounds
-
Backprop-Q: Generalized Backpropagation for Stochastic Computation Graphs
-
Efficient Probabilistic Inference in the Quest for Physics Beyond the Standard Model
-
CNN Features off-the-shelf: an Astounding Baseline for Recognition
-
Multitask Reinforcement Learning for Zero-shot Generalization with Subtask Dependencies
-
An Opinionated Introduction to AutoML and Neural Architecture Search
-
ActiveStereoNet: End-to-End Self-Supervised Learning for Active Stereo Systems
-
IntroVAE: Introspective Variational Autoencoders for Photographic Image Synthesis
-
Variance Networks: When Expectation Does Not Meet Your Expectations
-
Task-Driven Convolutional Recurrent Models of the Visual System
-
Co-Training of Audio and Video Representations from Self-Supervised Temporal Synchronization
-
Dynamic Integration of Background Knowledge in Neural NLU Systems
-
The relativistic discriminator: a key element missing from standard GAN
-
Accurate Uncertainties for Deep Learning Using Calibrated Regression
-
ModaNet: A Large-Scale Street Fashion Dataset with Polygon Annotations
-
The challenge of realistic music generation: modelling raw audio at scale
-
Every Pixel Counts: Unsupervised Geometry Learning with Holistic 3D Motion Understanding
-
Differentiable Compositional Kernel Learning for Gaussian Processes
-
Global Pose Estimation with an Attention-based Recurrent Network
-
Fighting Fake News: Image Splice Detection via Learned Self-Consistency
-
Spectral Inference Networks: Unifying Spectral Methods With Deep Learning
-
Granger-causal Attentive Mixtures of Experts: Learning Important Features with Neural Networks
-
Relational inductive biases, deep learning, and graph networks
-
Multi-Resolution 3D Convolutional Neural Networks for Object Recognition
-
Structure from noise: Mental errors yield abstract represen- tations of events
-
Learning to Extract Coherent Summary via Deep Reinforcement Learning
-
MolGAN: An implicit generative model for small molecular graphs
-
Variational Inference for Data-Efficient Model Learning in POMDPs
-
Learning a Prior over Intent via Meta-Inverse Reinforcement Learning
-
Aggregated Residual Transformations for Deep Neural Networks
-
Visceral Machines: Reinforcement Learning with Intrinsic Rewards that Mimic the Human Nervous System
-
AlphaX: eXploring Neural Architectures with Deep Neural Networks and Monte Carlo Tree Search
-
Progress & Compress: A scalable framework for continual learning
-
On the Practical Computational Power of Finite Precision RNNs for Language Recognition
-
Learning Intrinsic Image Decomposition from Watching the World
-
Pixel2Mesh: Generating 3D Mesh Models from Single RGB Images
-
Pedestrian-Synthesis-GAN: Generating Pedestrian Data in Real Scene and Beyond
-
Learning Intrinsic Image Decomposition from Watching the World
-
Time Series Epenthesis: Clustering Time Series Streams Requires Ignoring Some Data
-
Graph2Seq: Graph to Sequence Learning with Attention-based Neural Networks
-
Learning to Separate Object Sounds by Watching Unlabeled Video
-
Bayesian Robust Tensor Factorization for Incomplete Multiway Data
-
Quantization Mimic: Towards Very Tiny CNN for Object Detection
-
Efficient Convolutional Network Learning using Parametric Log based Dual-Tree Wavelet ScatterNet
-
Zero-shot Recognition via Semantic Embeddings and Knowledge Graphs
-
An Empirical Evaluation of Generic Convolutional and Recurrent Networks for Sequence Modeling
-
Realistic Evaluation of Deep Semi-Supervised Learning Algorithms
-
MergeNet: A Deep Net Architecture for Small Obstacle Discovery
-
Harmonic Networks: Deep Translation and Rotation Equivariance
-
ST-GAN: Spatial Transformer Generative Adversarial Networks for Image Compositing
-
Detecting Malicious PowerShell Commands using Deep Neural Networks
-
Regularizing Deep Networks by Modeling and Predicting Label Structure
-
Generative Adversarial Networks for Extreme Learned Image Compression
-
Concept2vec: Metrics for Evaluating Quality of Embeddings for Ontological Concepts
-
Differentiable plasticity: training plastic neural networks with backpropagation
-
sense2vec - A Fast and Accurate Method for Word Sense Disambiguation In Neural Word Embeddings
-
L4: Practical loss-based stepsize adaptation for deep learning
-
OverFeat: Integrated Recognition, Localization and Detection using Convolutional Networks
-
Complex-YOLO: An Euler-Region-Proposal for Real-time 3D Object Detection on Point Clouds
-
Text2Shape: Generating Shapes from Natural Language by Learning Joint Embeddings
-
Deep Binaries: Encoding Semantic-Rich Cues for Efficient Textual-Visual Cross Retrieval
-
Deep k-Nearest Neighbors: Towards Confident, Interpretable and Robust Deep Learning
-
Simple random search provides a competitive approach to reinforcement learning
-
ChatPainter: Improving Text to Image Generation using Dialogue
-
Deep Metric Learning via Lifted Structured Feature Embedding
-
Hard Negative Mining for Metric Learning Based Zero-Shot Classification
-
Triplet-based Deep Similarity Learning for Person Re-Identification
-
Robust Monitoring of Time Series with Application to Fraud Detection
-
Neural Machine Translation by Jointly Learning to Align and Translate
-
Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translation
-
Man is to Computer Programmer as Woman is to Homemaker? Debiasing Word Embeddings
-
Learning a Parametric Embedding by Preserving Local Structure
-
Explanations based on the Missing: Towards Contrastive Explanations with Pertinent Negatives
-
Tensor Comprehensions: Framework-Agnostic High-Performance Machine Learning Abstractions
-
End-to-end Deep Image Reconstruction From Human Brain Activity
-
TSSD: Temporal Single-Shot Detector Based on Attention and LSTM for Robotic Intelligent Perception
-
Deep Multimodal Learning For Emotion Recognition In Spoken Language
-
On Characterizing the Capacity of Neural Networks using Algebraic Topology
-
Adversarial Examples that Fool both Human and Computer Vision
-
Batch Renormalization: Towards Reducing Minibatch Dependence in Batch-Normalized Models
-
Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
-
Supervised Dimensionality Reduction via Distance Correlation Maximization
-
Faster gaze prediction with dense networks and Fisher pruning
-
IMPALA: Scalable Distributed Deep-RL with Importance Weighted Actor-Learner Architectures
-
Deep Temporal Clustering : Fully Unsupervised Learning of Time-Domain Features
-
Regularized Evolution for Image Classifier Architecture Search
-
Speed/accuracy trade-offs for modern convolutional object detectors
-
Towards AI-Complete Question Answering: A Set of Prerequisite Toy Tasks
-
Audio Adversarial Examples: Targeted Attacks on Speech-to-Text
-
ComboGAN: Unrestrained Scalability for Image Domain Translation
-
Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks
-
Mastering Chess and Shogi by Self-Play with a General Reinforcement Learning Algorithm
-
High-Resolution Image Synthesis and Semantic Manipulation with Conditional GANs
-
Understanding deep learning requires rethinking generalization
-
VoxelNet: End-to-End Learning for Point Cloud Based 3D Object Detection
-
DeepFace: Closing the Gap to Human-Level Performance in Face Verification
-
Full-Resolution Residual Networks for Semantic Segmentation in Street Scenes
-
Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
-
Rich feature hierarchies for accurate object detection and semantic segmentation
-
Stereo Matching by Training a Convolutional Neural Network to Compare Image Patches
-
Semantic Segmentation of Earth Observation Data Using Multimodal and Multi-scale Deep Networks
-
3D Object Proposals using Stereo Imagery for Accurate Object Class Detection
-
Wider or Deeper: Revisiting the ResNet Model for Visual Recognition
-
The Cityscapes Dataset for Semantic Urban Scene Understanding
-
Superpixel Convolutional Networks Using Bilateral Inceptions
-
Aggressive Deep Driving: Combining Convolutional Neural Networks and Model Predictive Control
-
Multi-View 3D Object Detection Network for Autonomous Driving
-
High-Resolution Semantic Labeling with Convolutional Neural Networks
-
Instance-aware Semantic Segmentation via Multi-task Network Cascades
-
Vehicle Detection from 3D Lidar Using Fully Convolutional Network
-
OctNet: Learning Deep 3D Representations at High Resolutions
-
EnhanceNet: Single Image Super-Resolution Through Automated Texture Synthesis
-
Hierarchical Representations for Efficient Architecture Search
-
Improving Deep Learning by Inverse Square Root Linear Units (ISRLUs)
-
Long Short-Term Memory as a Dynamically Computed Element-wise Weighted Sum
-
Soft-NMS -- Improving Object Detection With One Line of Code
-
A Unified Multi-scale Deep Convolutional Neural Network for Fast Object Detection
-
SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation
-
Incremental Dense Semantic Stereo Fusion for Large-Scale Semantic Scene Reconstruction
-
StixelNet: A Deep Convolutional Network for Obstacle Detection and Road Segmentation
-
Semantic Image Segmentation with Deep Cconvolutional Nets and Fully Connected CRFS
-
A New Artificial Neural Network Approach in Solving Inverse Kinematics of Robotic Arm (Denso VP6242)
-
Auto-Conditioned LSTM Network for Extended Complex Human Motion Synthesis
-
DeepXplore: Automated Whitebox Testing of Deep Learning Systems
-
Real-time grasp detection using convolutional neural networks
-
Learning real manipulation tasks from virtual demonstrations using LSTM
-
A Neural Network-Based Approach for Trajectory Planning in Robot–Human Handover Tasks
-
SE3-Nets: Learning Rigid Body Motion using Deep Neural Networks
-
UnDeepVO: Monocular Visual Odometry through Unsupervised Deep Learning
-
Supersizing Self-supervision: Learning to Grasp from 50K Tries and 700 Robot Hours
-
3D Simulated Robot Manipulation Using Deep Reinforcement Learning
-
Deep-learning in Mobile Robotics - from Perception to Control Systems: A Survey on Why and Why not
-
Leveraging Deep Reinforcement Learning for Reaching Robotic Tasks
-
Imagenet classification with deep convolutional neural networks
-
Supervision via Competition: Robot Adversaries for Learning Tasks
-
Dermatologist-level classification of skin cancer with deep neural networks
-
Understanding image representations by measuring their equivariance and equivalence
-
Batch normalization: Accelerating deep network training by reducing internal covariate shift
-
Dropout: A simple way to prevent neural networks from overfitting
-
Delving deep into rectifiers: Surpassing human-level performance on imagenet classification
-
Improving neural networks by preventing co-adaptation of feature detectors
-
PoseNet: A Convolutional Network for Real-Time 6-DOF Camera Relocalization
-
Systematic Testing of Convolutional Neural Networks for Autonomous Driving
-
Deep Visual-Semantic Alignments for Generating Image Descriptions
-
Very Deep Convolutional Networks for Large-scale Image Recognition
-
Clinical Intervention Prediction and Understanding with Deep Neural Networks
-
Connecting Images and Natural Language (Andrej Karpathy's Dissertation)
-
Improving Neural Networks by Preventing Co-adaptation of Feature Detectors
-
U-Net: Convolutional Networks for Biomedical Image Segmentation
-
Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
-
Modeling Smooth Backgrounds & Generic Localized Signals with Gaussian Processes (Kyle)
-
Adversarial Variational Optimization of Non-Differentiable Simulators (Kyle)
-
Rules of Machine Learning: Best Practices for ML Engineering
-
Fast Near-Duplicate Image Search using Locality Sensitive Hashing
-
Loc2Vec: Learning location embeddings with triplet-loss networks
-
Using Word2Vec for Better Embeddings of Categorical Features
-
Stochastic Weight Averaging — a New Way to Get State of the Art Results in Deep Learning
-
wav2letter: A Facebook AI Research(FAIR) Automatic Speech Recognition Toolkit
-
Using Evolutionary AutoML to Discover Neural Network Architectures
-
Using Tensorflow Object Detection to do Pixel Wise Classification
-
Understanding GANs through a statistical divergence perspective
-
Understanding and optimizing GANs (Going back to first principles)
-
Understanding Learning Rates and How It Improves Performance in Deep Learning
-
Only Numpy: Implementing Highway Network, OOP approach with Mini Batch with Interactive Code
-
Only Numpy: NIPS 2017 - Implementing Dilated Recurrent Neural Networks with Interactive Code
-
Annotating Large Datasets with the TensorFlow Object Detection API
-
Deep Learning Is Not Good Enough, We Need Bayesian Deep Learning for Safe AI
-
SBNet: Leveraging Activation Block Sparsity for Speeding up Convolutional Neural Networks
-
A set of Deep Reinforcement Learning Agents implemented in Tensorflow
-
Deep Learning Adversarial Examples – Clarifying Misconceptions
-
How to Define an Encoder-Decoder Sequence-to-Sequence Model for Neural Machine Translation in Keras
-
Feature Visualization - How neural networks build up their understanding of images
-
How to Develop a Character-Based Neural Language Model in Keras
-
How-To: Multi-GPU training with Keras, Python, and deep learning
-
Best Practices for Document Classification with Deep Learning
-
Leonardo Araujo dos Santos' Artificial Intelligence and Deep Learning GitBook
-
A quick complete tutorial to save and restore Tensorflow models
-
The Mathematics of 2048: Optimal Play with Markov Decision Processes
-
Embed, encode, attend, predict: The new deep learning formula for state-of-the-art NLP models
-
Army develops face recognition technology that works in the dark
-
Generative Adversarial Networks for Extreme Learned Image Compression
-
Splash of Color: Instance Segmentation with Mask R-CNN and TensorFlow
-
Machine Learning is Fun Part 6: How to do Speech Recognition with Deep Learning
-
Engineering Extreme Event Forecasting at Uber with Recurrent Neural Networks
-
Releasing “Supervisely Person” dataset for teaching machines to segment humans
-
Only Numpy: Implementing Mini VGG (VGG 7) and SoftMax Layer with Interactive Code
-
An overview of word embeddings and their connection to distributional semantic models
-
NVIDIA Researchers Showcase Major Advances in Deep Learning at NIPS
-
GAN Playground - Explore Generative Adversarial Nets in your Browser
-
GAN Playground - Explore Generative Adversarial Nets in your Browser
-
Teaching a car to avoid obstacles using Reinforcement Learning
-
Creating Photorealistic Images with Neural Networks and a Gameboy Camera
-
Will Roscoe's Lane Following Autopilot with Keras & Tensorflow