Awesome Open Source
Awesome Open Source

Awesome AI books

Some awesome AI related books and pdfs for downloading and learning.


This repo only used for learning, do not use in business.

Welcome for providing great books in this repo or tell me which great book you need and I will try to append it in this repo, any idea you can create issue or PR here.

Due to github Large file storage limition, all books pdf stored in Yandex.Disk.

Some often used Mathematic Symbols can refer this page


Organization with papers/researchs

Training ground

  • OpenAI Gym: A toolkit for developing and comparing reinforcement learning algorithms. (Can play with Atari, Box2d, MuJoCo etc...)
  • DeepMind Pysc2: StarCraft II Learning Environment.
  • Procgen: Procgen Benchmark: Procedurally-Generated Game-Like Gym-Environments.
  • TorchCraftAI: A bot platform for machine learning research on StarCraft®: Brood War®
  • Valve Dota2: Dota2 game acessing api. (CN doc)
  • Mario AI Framework: A Mario AI framework for using AI methods.
  • Google Dopamine: Dopamine is a research framework for fast prototyping of reinforcement learning algorithms
  • TextWorld: Microsoft - A learning environment sandbox for training and testing reinforcement learning (RL) agents on text-based games.
  • Mini Grid: Minimalistic gridworld environment for OpenAI Gym
  • MAgent: A Platform for Many-agent Reinforcement Learning
  • XWorld: A C++/Python simulator package for reinforcement learning
  • Neural MMO: A Massively Multiagent Game Environment
  • MinAtar: MinAtar is a testbed for AI agents which implements miniaturized version of several Atari 2600 games.
  • craft-env: CraftEnv is a 2D crafting environment
  • gym-sokoban: Sokoban is Japanese for warehouse keeper and a traditional video game
  • Pommerman Playground hosts Pommerman, a clone of Bomberman built for AI research.
  • gym-miniworld MiniWorld is a minimalistic 3D interior environment simulator for reinforcement learning & robotics research
  • vizdoomgym OpenAI Gym wrapper for ViZDoom (A Doom-based AI Research Platform for Reinforcement Learning from Raw Visual Information) enviroments.
  • ddz-ai 以孤立语假设和宽度优先搜索为基础,构建了一种多通道堆叠注意力Transformer结构的斗地主ai


Introductory theory and get start


Data mining

Machine Learning

Deep Learning


Quantum with AI

Libs With Online Books

  • Reinforcement Learning

    • A3C - Google DeepMind Asynchronous Advantage Actor-Critic algorithm
    • Q-Learning SARSA DQN DDQN - Q-Learning is a value-based Reinforcement Learning algorithm
    • DDPG - Deep Deterministic Policy Gradient,
    • Large-Scale Curiosity - Large-Scale Study of Curiosity-Driven Learning
    • PPO - OpenAI Proximal Policy Optimization Algorithms
    • RND - OpenAI Random Network Distillation, an exploration bonus for deep reinforcement learning method.
    • VIME - OpenAI Variational Information Maximizing Exploration
    • DQV - Deep Quality-Value (DQV) Learning
    • ERL - Evolution-Guided Policy Gradient in Reinforcement Learning
    • MF Multi-Agent RL - Mean Field Multi-Agent Reinforcement Learning. (this paper include MF-Q and MF-AC)
    • MAAC - Actor-Attention-Critic for Multi-Agent Reinforcement Learning
  • Feature Selection

  • Machine Learning

    • Xgboost (Python, R, JVM, Julia, CLI) - Xgboost lib's document.
    • LightGBM (Python, R, CLI) - Microsoft lightGBM lib's features document.
    • CatBoost (Python, R, CLI) - Yandex Catboost lib's key algorithm pdf papper.
    • StackNet (Java, CLI) - Some model stacking algorithms implemented in this lib.
    • RGF - Learning Nonlinear Functions Using Regularized Greedy Forest (multi-core implementation FastRGF)
    • FM, FastFM, FFM, XDeepFM - Factorization Machines and some extended Algorithms
  • Deep Learning

    • GNN Papers - Must-read papers on graph neural networks (GNN)
    • EfficientNet - Rethinking Model Scaling for Convolutional Neural Networks
    • DenseNet - Densely Connected Convolutional Networks
  • NLP

    • XLNet - repo XLNet: Generalized Autoregressive Pretraining for Language Understanding
    • BERT - Pre-training of Deep Bidirectional Transformers for Language Understanding
    • GPT-3 - Language Models are Few-Shot Learners
  • CV

    • Fast R-CNN - Fast Region-based Convolutional Network method (Fast R-CNN) for object detection
    • Mask R-CNN - Mask R-CNN, extends Faster R-CNN by adding a branch for predicting an object mask in parallel with the existing branch for bounding box recognition.
    • GQN - DeepMind Generative Query Network, Neural scene representation and rendering
  • Meta Learning

    • MAML - Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
  • Transfer Learning

    • GCN - Zero-shot Recognition via Semantic Embeddings and Knowledge Graphs
  • Auto ML

    • Model Search (Python) - Google Model search (MS) is a framework that implements AutoML algorithms for model architecture search at scale.
    • TPOT (Python) - TPOT is a lib for AutoML.
    • Auto-sklearn (Python) - auto-sklearn is an automated machine learning toolkit and a drop-in replacement for a scikit-learn estimator
    • Auto-Keras (Python) - Auto-Keras is an open source software library for automated machine learning (AutoML). It is developed by DATA Lab
    • TransmogrifAI (JVM) - TransmogrifAI (pronounced trăns-mŏgˈrə-fī) is an AutoML library written in Scala that runs on top of Spark
    • Auto-WEKAA - Provides automatic selection of models and hyperparameters for WEKA.
    • MLBox (Python) - MLBox is a powerful Automated Machine Learning python library
  • Pipeline Training

    • ZenML (Python) - ZenML is built for ML practitioners who are ramping up their ML workflows towards production
  • Dimensionality Reduction

    • t-SNE (Non-linear/Non-params) - T-distributed Stochastic Neighbor Embedding (t-SNE) is a machine learning algorithm for visualization
    • PCA (Linear) - Principal component analysis
    • LDA (Linear) - Linear Discriminant Analysis
    • LLE (Non-linear) - Locally linear embedding
    • Laplacian Eigenmaps - Laplacian Eigenmaps for Dimensionality Reduction and Data Representation
    • Sammon Mapping (Non-linear) - Sammon mapping is designed to minimise the differences between corresponding inter-point distances in the two spaces

Distributed training

  • Horovod - Horovod is a distributed training framework for TensorFlow, Keras, PyTorch, and MXNet. The goal of Horovod is to make distributed Deep Learning fast and easy to use.


Code Contributors

This project exists thanks to all the people who contribute. [Contribute].

Financial Contributors

Become a financial contributor and help us sustain our community. [Contribute]



Support this project with your organization. Your logo will show up here with a link to your website. [Contribute]

Get A Weekly Email With Trending Projects For These Topics
No Spam. Unsubscribe easily at any time.
jupyter-notebook (6,156
deep-learning (3,906
machine-learning (3,567
artificial-intelligence (627
reinforcement-learning (551
ai (463
algorithms (454
pdf (442
learning (357
data-mining (168
mathematics (153
books (125
playground (89
quantum-computing (83
reading (22
quantum-algorithms (19