[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"tool-dair-ai--ML-YouTube-Courses":3,"similar-dair-ai--ML-YouTube-Courses":80},{"id":4,"github_repo":5,"name":6,"description_en":7,"description_zh":8,"ai_summary_zh":8,"readme_en":9,"readme_zh":10,"quickstart_zh":11,"use_case_zh":12,"hero_image_url":13,"owner_login":14,"owner_name":15,"owner_avatar_url":16,"owner_bio":17,"owner_company":18,"owner_location":18,"owner_email":18,"owner_twitter":19,"owner_website":20,"owner_url":21,"languages":18,"stars":22,"forks":23,"last_commit_at":24,"license":25,"difficulty_score":26,"env_os":27,"env_gpu":28,"env_ram":28,"env_deps":29,"category_tags":32,"github_topics":39,"view_count":46,"oss_zip_url":18,"oss_zip_packed_at":18,"status":47,"created_at":48,"updated_at":49,"faqs":50,"releases":79},3011,"dair-ai\u002FML-YouTube-Courses","ML-YouTube-Courses","📺 Discover the latest machine learning \u002F AI courses on YouTube.","ML-YouTube-Courses 是一个由 DAIR.AI 维护的开源项目，致力于汇集并整理 YouTube 上最新、最优质的机器学习与人工智能课程资源。面对网络上分散且质量参差不齐的教学视频，它通过系统化的分类索引，帮助用户快速定位从基础理论到前沿应用的核心内容。\n\n该资源库覆盖了机器学习、深度学习、科学计算、大语言模型工程（LLMOps）、自然语言处理及计算机视觉等多个关键领域。其中不仅收录了斯坦福、MIT、加州理工等顶尖高校的经典课程（如 CS229、CS231N），还紧跟技术潮流，纳入了关于 Transformer 架构、生成式 AI 评估、LangChain 应用开发等实战教程。\n\n无论是希望夯实理论基础的研究人员、需要掌握最新工程落地的开发者，还是对 AI 充满好奇的学习者，都能在此找到适合的学习路径。其独特亮点在于将原本零散的视频资源整合为结构清晰的知识体系，并持续更新以反映行业动态，让用户无需耗费大量时间搜索筛选，即可享受系统化的高质量开放教育体验。","# 📺 ML YouTube Courses\n\nAt DAIR.AI we ❤️ open AI education. In this repo, we index and organize some of the best and most recent machine learning courses available on YouTube.\n\n**Machine Learning**\n\n- [Caltech CS156: Learning from Data](#caltech-cs156-learning-from-data)\n- [Stanford CS229: Machine Learning](#stanford-cs229-machine-learning)\n- [Making Friends with Machine Learning](#making-friends-with-machine-learning)\n- [Applied Machine Learning](#applied-machine-learning)\n- [Introduction to Machine Learning (Tübingen)](#introduction-to-machine-learning-Tübingen)\n- [Machine Learning Lecture (Stefan Harmeling)](#machine-learning-lecture-stefan-harmeling)\n- [Statistical Machine Learning (Tübingen)](#statistical-machine-learning-Tübingen)\n- [Probabilistic Machine Learning](#probabilistic-machine-learning)\n- [MIT 6.S897: Machine Learning for Healthcare (2019)](#mit-6s897-machine-learning-for-healthcare-2019)\n\n**Deep Learning**\n\n- [Neural Networks: Zero to Hero](#neural-networks-zero-to-hero-by-andrej-karpathy)\n- [MIT: Deep Learning for Art, Aesthetics, and Creativity](#mit-deep-learning-for-art-aesthetics-and-creativity)\n- [Stanford CS230: Deep Learning (2018)](#stanford-cs230-deep-learning-2018)\n- [Introduction to Deep Learning (MIT)](#introduction-to-deep-learning)\n- [CMU Introduction to Deep Learning (11-785)](#cmu-introduction-to-deep-learning-11-785)\n- [Deep Learning: CS 182](#deep-learning-cs-182)\n- [Deep Unsupervised Learning](#deep-unsupervised-learning)\n- [NYU Deep Learning SP21](#nyu-deep-learning-sp21)\n- [Foundation Models](#foundation-models)\n- [Deep Learning (Tübingen)](#deep-learning-Tübingen)\n\n**Scientific Machine Learning**\n\n- [Parallel Computing and Scientific Machine Learning](#parallel-computing-and-scientific-machine-learning)\n\n**Practical Machine Learning**\n\n- [LLMOps: Building Real-World Applications With Large Language Models](#llmops-building-real-world-applications-with-large-language-models)\n- [Evaluating and Debugging Generative AI](#evaluating-and-debugging-generative-ai)\n- [ChatGPT Prompt Engineering for Developers](#chatgpt-prompt-engineering-for-developers)\n- [LangChain for LLM Application Development](#langchain-for-llm-application-development)\n- [LangChain: Chat with Your Data](#langchain-chat-with-your-data)\n- [Building Systems with the ChatGPT API](#building-systems-with-the-chatgpt-api)\n- [LangChain & Vector Databases in Production](#langchain--vector-databases-in-production)\n- [Building LLM-Powered Apps](#building-llm-powered-apps)\n- [Full Stack LLM Bootcamp](#full-stack-llm-bootcamp)\n- [Full Stack Deep Learning](#full-stack-deep-learning)\n- [Practical Deep Learning for Coders](#practical-deep-learning-for-coders)\n- [Stanford MLSys Seminars](#stanford-mlsys-seminars)\n- [Machine Learning Engineering for Production (MLOps)](#machine-learning-engineering-for-production-mlops)\n- [MIT Introduction to Data-Centric AI](#mit-introduction-to-data-centric-ai)\n\n**Natural Language Processing**\n\n- [XCS224U: Natural Language Understanding (2023)](#xcs224u-natural-language-understanding-2023)\n- [Stanford CS25 - Transformers United](#stanford-cs25---transformers-united)\n- [NLP Course (Hugging Face)](#nlp-course-hugging-face)\n- [CS224N: Natural Language Processing with Deep Learning](#cs224n-natural-language-processing-with-deep-learning)\n- [CMU Neural Networks for NLP](#cmu-neural-networks-for-nlp)\n- [CS224U: Natural Language Understanding](#cs224u-natural-language-understanding)\n- [CMU Advanced NLP 2021\u002F2022\u002F2024](#cmu-advanced-nlp)\n- [Multilingual NLP](#multilingual-nlp)\n- [Advanced NLP](#advanced-nlp)\n\n**Computer Vision**\n\n- [CS231N: Convolutional Neural Networks for Visual Recognition](#cs231n-convolutional-neural-networks-for-visual-recognition)\n- [Deep Learning for Computer Vision](#deep-learning-for-computer-vision)\n- [Deep Learning for Computer Vision (DL4CV)](#deep-learning-for-computer-vision-dl4cv)\n- [Deep Learning for Computer Vision (neuralearn.ai)](#deep-learning-for-computer-vision-neuralearnai)\n\n**Reinforcement Learning**\n\n- [Deep Reinforcement Learning](#deep-reinforcement-learning)\n- [Reinforcement Learning Lecture Series (DeepMind)](#reinforcement-learning-lecture-series-deepmind)\n- [Reinforcement Learning (Polytechnique Montreal, Fall 2021)](#reinforcement-learning-polytechnique-montreal-fall-2021)\n- [Foundations of Deep RL](#foundations-of-deep-rl)\n- [Stanford CS234: Reinforcement Learning](#stanford-cs234-reinforcement-learning)\n\n**Graph Machine Learning**\n\n- [Machine Learning with Graphs (Stanford)](#machine-learning-with-graphs-stanford)\n- [AMMI Geometric Deep Learning Course](#ammi-geometric-deep-learning-course)\n\n**Multi-Task Learning**\n\n- [Multi-Task and Meta-Learning (Stanford)](#stanford-cs330-deep-multi-task-and-meta-learning)\n\n**Others**\n\n- [MIT Deep Learning in Life Sciences](#mit-deep-learning-in-life-sciences)\n- [Self-Driving Cars (Tübingen)](#self-driving-cars-Tübingen)\n- [Advanced Robotics (Berkeley)](#advanced-robotics-uc-berkeley)\n\n---\n\n## Caltech CS156: Learning from Data\n\nAn introductory course in machine learning that covers the basic theory, algorithms, and applications.\n\n- Lecture 1: The Learning Problem\n- Lecture 2: Is Learning Feasible?\n- Lecture 3: The Linear Model I\n- Lecture 4: Error and Noise\n- Lecture 5: Training versus Testing\n- Lecture 6: Theory of Generalization\n- Lecture 7: The VC Dimension\n- Lecture 8: Bias-Variance Tradeoff\n- Lecture 9: The Linear Model II\n- Lecture 10: Neural Networks\n- Lecture 11: Overfitting\n- Lecture 12: Regularization\n- Lecture 13: Validation\n- Lecture 14: Support Vector Machines\n- Lecture 15: Kernel Methods\n- Lecture 16: Radial Basis Functions\n- Lecture 17: Three Learning Principles\n- Lecture 18: Epilogue\n\n🔗 [Link to Course](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLD63A284B7615313A)\n\n## Stanford CS229: Machine Learning\n\nTo learn some of the basics of ML:\n\n- Linear Regression and Gradient Descent\n- Logistic Regression\n- Naive Bayes\n- SVMs\n- Kernels\n- Decision Trees\n- Introduction to Neural Networks\n- Debugging ML Models\n  ...\n\n🔗 [Link to Course](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLoROMvodv4rMiGQp3WXShtMGgzqpfVfbU)\n\n## Making Friends with Machine Learning\n\nA series of mini lectures covering various introductory topics in ML:\n\n- Explainability in AI\n- Classification vs. Regression\n- Precession vs. Recall\n- Statistical Significance\n- Clustering and K-means\n- Ensemble models\n  ...\n\n🔗 [Link to Course](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLRKtJ4IpxJpDxl0NTvNYQWKCYzHNuy2xG)\n\n## Neural Networks: Zero to Hero (by Andrej Karpathy)\n\nCourse providing an in-depth overview of neural networks.\n\n- Backpropagation\n- Spelled-out intro to Language Modeling\n- Activation and Gradients\n- Becoming a Backprop Ninja\n\n🔗 [Link to Course](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLAqhIrjkxbuWI23v9cThsA9GvCAUhRvKZ)\n\n## MIT: Deep Learning for Art, Aesthetics, and Creativity\n\nCovers the application of deep learning for art, aesthetics, and creativity.\n\n- Nostalgia -> Art -> Creativity -> Evolution as Data + Direction\n- Efficient GANs\n- Explorations in AI for Creativity\n- Neural Abstractions\n- Easy 3D Content Creation with Consistent Neural Fields\n  ...\n\n🔗 [Link to Course](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLCpMvp7ftsnIbNwRnQJbDNRqO6qiN3EyH)\n\n## Stanford CS230: Deep Learning (2018)\n\nCovers the foundations of deep learning, how to build different neural networks(CNNs, RNNs, LSTMs, etc...), how to lead machine learning projects, and career advice for deep learning practitioners.\n\n- Deep Learning Intuition\n- Adversarial examples - GANs\n- Full-cycle of a Deep Learning Project\n- AI and Healthcare\n- Deep Learning Strategy\n- Interpretability of Neural Networks\n- Career Advice and Reading Research Papers\n- Deep Reinforcement Learning\n\n🔗 [Link to Course](https:\u002F\u002Fyoutube.com\u002Fplaylist?list=PLoROMvodv4rOABXSygHTsbvUz4G_YQhOb) 🔗 [Link to Materials](https:\u002F\u002Fcs230.stanford.edu\u002Fsyllabus\u002F)\n\n## Applied Machine Learning\n\nTo learn some of the most widely used techniques in ML:\n\n- Optimization and Calculus\n- Overfitting and Underfitting\n- Regularization\n- Monte Carlo Estimation\n- Maximum Likelihood Learning\n- Nearest Neighbours\n- ...\n\n🔗 [Link to Course](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PL2UML_KCiC0UlY7iCQDSiGDMovaupqc83)\n\n## Introduction to Machine Learning (Tübingen)\n\nThe course serves as a basic introduction to machine learning and covers key concepts in regression, classification, optimization, regularization, clustering, and dimensionality reduction.\n\n- Linear regression\n- Logistic regression\n- Regularization\n- Boosting\n- Neural networks\n- PCA\n- Clustering\n- ...\n\n🔗 [Link to Course](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PL05umP7R6ij35ShKLDqccJSDntugY4FQT)\n\n## Machine Learning Lecture (Stefan Harmeling)\n\nCovers many fundamental ML concepts:\n\n- Bayes rule\n- From logic to probabilities\n- Distributions\n- Matrix Differential Calculus\n- PCA\n- K-means and EM\n- Causality\n- Gaussian Processes\n- ...\n\n🔗 [Link to Course](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLzrCXlf6ypbxS5OYOY3EN_0u2fDuIT6Gt)\n\n## Statistical Machine Learning (Tübingen)\n\nThe course covers the standard paradigms and algorithms in statistical machine learning.\n\n- KNN\n- Bayesian decision theory\n- Convex optimization\n- Linear and ridge regression\n- Logistic regression\n- SVM\n- Random Forests\n- Boosting\n- PCA\n- Clustering\n- ...\n\n🔗 [Link to Course](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PL05umP7R6ij2XCvrRzLokX6EoHWaGA2cC)\n\n## Practical Deep Learning for Coders\n\nThis course covers topics such as how to:\n\n- Build and train deep learning models for computer vision, natural language processing, tabular analysis, and collaborative filtering problems\n- Create random forests and regression models\n- Deploy models\n- Use PyTorch, the world’s fastest growing deep learning software, plus popular libraries like fastai and Hugging Face\n- Foundations and Deep Dive to Diffusion Models\n- ...\n\n🔗 [Link to Course - Part 1](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLfYUBJiXbdtSvpQjSnJJ_PmDQB_VyT5iU)\n\n🔗 [Link to Course - Part 2](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=_7rMfsA24Ls&ab_channel=JeremyHoward)\n\n## Stanford MLSys Seminars\n\nA seminar series on all sorts of topics related to building machine learning systems.\n\n🔗 [Link to Lectures](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLSrTvUm384I9PV10koj_cqit9OfbJXEkq)\n\n## Machine Learning Engineering for Production (MLOps)\n\nSpecialization course on MLOPs by Andrew Ng.\n\n🔗 [Link to Lectures](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLkDaE6sCZn6GMoA0wbpJLi3t34Gd8l0aK)\n\n## MIT Introduction to Data-Centric AI\n\nCovers the emerging science of Data-Centric AI (DCAI) that studies techniques to improve datasets, which is often the best way to improve performance in practical ML applications. Topics include:\n\n- Data-Centric AI vs. Model-Centric AI\n- Label Errors\n- Dataset Creation and Curation\n- Data-centric Evaluation of ML Models\n- Class Imbalance, Outliers, and Distribution Shift\n- ...\n\n🔗 [Course Website](https:\u002F\u002Fdcai.csail.mit.edu\u002F)\n\n🔗 [Lecture Videos](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=ayzOzZGHZy4&list=PLnSYPjg2dHQKdig0vVbN-ZnEU0yNJ1mo5)\n\n🔗 [Lab Assignments](https:\u002F\u002Fgithub.com\u002Fdcai-course\u002Fdcai-lab)\n\n## Machine Learning with Graphs (Stanford)\n\nTo learn some of the latest graph techniques in machine learning:\n\n- PageRank\n- Matrix Factorizing\n- Node Embeddings\n- Graph Neural Networks\n- Knowledge Graphs\n- Deep Generative Models for Graphs\n- ...\n\n🔗 [Link to Course](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLoROMvodv4rPLKxIpqhjhPgdQy7imNkDn)\n\n## Probabilistic Machine Learning\n\nTo learn the probabilistic paradigm of ML:\n\n- Reasoning about uncertainty\n- Continuous Variables\n- Sampling\n- Markov Chain Monte Carlo\n- Gaussian Distributions\n- Graphical Models\n- Tuning Inference Algorithms\n- ...\n\n🔗 [Link to Course](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PL05umP7R6ij2YE8rRJSb-olDNbntAQ_Bx)\n\n## MIT 6.S897: Machine Learning for Healthcare (2019)\n\nThis course introduces students to machine learning in healthcare, including the nature of clinical data and the use of machine learning for risk stratification, disease progression modeling, precision medicine, diagnosis, subtype discovery, and improving clinical workflows.\n\n🔗 [Link to Course](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLUl4u3cNGP60B0PQXVQyGNdCyCTDU1Q5j)\n\n## Introduction to Deep Learning\n\nTo learn some of the fundamentals of deep learning:\n\n- Introduction to Deep Learning\n\n🔗 [Link to Course](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLtBw6njQRU-rwp5__7C0oIVt26ZgjG9NI)\n\n## CMU Introduction to Deep Learning (11-785)\n\nThe course starts off gradually from MLPs (Multi Layer Perceptrons) and then progresses into concepts like attention\nand sequence-to-sequence models.\n\n🔗 [Link to Course](https:\u002F\u002Fdeeplearning.cs.cmu.edu\u002FF22\u002Findex.html) \\\n🔗 [Lectures](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLp-0K3kfddPxRmjgjm0P1WT6H-gTqE8j9) \\\n🔗 [Tutorials\u002FRecitations](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLp-0K3kfddPz8WXg8RqH0sEN6X2L65HUZ)\n\n## Deep Learning: CS 182\n\nTo learn some of the widely used techniques in deep learning:\n\n- Machine Learning Basics\n- Error Analysis\n- Optimization\n- Backpropagation\n- Initialization\n- Batch Normalization\n- Style transfer\n- Imitation Learning\n- ...\n\n🔗 [Link to Course](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PL_iWQOsE6TfVmKkQHucjPAoRtIJYt8a5A)\n\n## Deep Unsupervised Learning\n\nTo learn the latest and most widely used techniques in deep unsupervised learning:\n\n- Autoregressive Models\n- Flow Models\n- Latent Variable Models\n- Self-supervised learning\n- Implicit Models\n- Compression\n- ...\n\n🔗 [Link to Course](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLwRJQ4m4UJjPiJP3691u-qWwPGVKzSlNP)\n\n## NYU Deep Learning SP21\n\nTo learn some of the advanced techniques in deep learning:\n\n- Neural Nets: rotation and squashing\n- Latent Variable Energy Based Models\n- Unsupervised Learning\n- Generative Adversarial Networks\n- Autoencoders\n- ...\n\n🔗 [Link to Course](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLLHTzKZzVU9e6xUfG10TkTWApKSZCzuBI)\n\n## Foundation Models\n\nTo learn about foundation models like GPT-3, CLIP, Flamingo, Codex, and DINO.\n\n🔗 [Link to Course](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PL9t0xVFP90GD8hox0KipBkJcLX_C3ja67)\n\n## Deep Learning (Tübingen)\n\nThis course introduces the practical and theoretical principles of deep neural networks.\n\n- Computation graphs\n- Activation functions and loss functions\n- Training, regularization and data augmentation\n- Basic and state-of-the-art deep neural network architectures including convolutional networks and graph neural networks\n- Deep generative models such as auto-encoders, variational auto-encoders and generative adversarial networks\n- ...\n\n🔗 [Link to Course](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PL05umP7R6ij3NTWIdtMbfvX7Z-4WEXRqD)\n\n## Parallel Computing and Scientific Machine Learning\n\n- The Basics of Scientific Simulators\n- Introduction to Parallel Computing\n- Continuous Dynamics\n- Inverse Problems and Differentiable Programming\n- Distributed Parallel Computing\n- Physics-Informed Neural Networks and Neural Differential Equations\n- Probabilistic Programming, AKA Bayesian Estimation on Programs\n- Globalizing the Understanding of Models\n\n🔗 [Link to Course](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLCAl7tjCwWyGjdzOOnlbGnVNZk0kB8VSa)\n\n## XCS224U: Natural Language Understanding (2023)\n\nThis course covers topics such as:\n\n- Contextual Word Representations\n- Information Retrieval\n- In-context learning\n- Behavioral Evaluation of NLU models\n- NLP Methods and Metrics\n- ...\n\n🔗 [Link to Course](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLoROMvodv4rOwvldxftJTmoR3kRcWkJBp)\n\n## Stanford CS25 - Transformers United\n\nThis course consists of lectures focused on Transformers, providing a deep dive and their applications\n\n- Introduction to Transformers\n- Transformers in Language: GPT-3, Codex\n- Applications in Vision\n- Transformers in RL & Universal\n  Compute Engines\n- Scaling transformers\n- Interpretability with transformers\n- ...\n\n🔗 [Link to Course](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLoROMvodv4rNiJRchCzutFw5ItR_Z27CM)\n\n## NLP Course (Hugging Face)\n\nLearn about different NLP concepts and how to apply language models and Transformers to NLP:\n\n- What is Transfer Learning?\n- BPE Tokenization\n- Batching inputs\n- Fine-tuning models\n- Text embeddings and semantic search\n- Model evaluation\n- ...\n\n🔗 [Link to Course](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLo2EIpI_JMQvWfQndUesu0nPBAtZ9gP1o)\n\n## CS224N: Natural Language Processing with Deep Learning\n\nTo learn the latest approaches for deep learning based NLP:\n\n- Dependency parsing\n- Language models and RNNs\n- Question Answering\n- Transformers and pretraining\n- Natural Language Generation\n- T5 and Large Language Models\n- Future of NLP\n- ...\n\n🔗 [Link to Course](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLoROMvodv4rOSH4v6133s9LFPRHjEmbmJ)\n\n## CMU Neural Networks for NLP\n\nTo learn the latest neural network based techniques for NLP:\n\n- Language Modeling\n- Efficiency tricks\n- Conditioned Generation\n- Structured Prediction\n- Model Interpretation\n- Advanced Search Algorithms\n- ...\n\n🔗 [Link to Course](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PL8PYTP1V4I8AkaHEJ7lOOrlex-pcxS-XV)\n\n## CS224U: Natural Language Understanding\n\nTo learn the latest concepts in natural language understanding:\n\n- Grounded Language Understanding\n- Relation Extraction\n- Natural Language Inference (NLI)\n- NLU and Neural Information Extraction\n- Adversarial testing\n- ...\n\n🔗 [Link to Course](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLoROMvodv4rPt5D0zs3YhbWSZA8Q_DyiJ)\n\n## CMU Advanced NLP\n\nTo learn:\n\n- Basics of modern NLP techniques\n- Multi-task, Multi-domain, multi-lingual learning\n- Prompting + Sequence-to-sequence pre-training\n- Interpreting and Debugging NLP Models\n- Learning from Knowledge-bases\n- Adversarial learning\n- ...\n\n🔗 [Link to 2021 Edition](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PL8PYTP1V4I8AYSXn_GKVgwXVluCT9chJ6)\n\n🔗 [Link to 2022 Edition](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PL8PYTP1V4I8D0UkqW2fEhgLrnlDW9QK7z)\n\n🔗 [Link to 2024 Edition](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PL8PYTP1V4I8DZprnWryM4nR8IZl1ZXDjg)\n\n## Multilingual NLP\n\nTo learn the latest concepts for doing multilingual NLP:\n\n- Typology\n- Words, Part of Speech, and Morphology\n- Advanced Text Classification\n- Machine Translation\n- Data Augmentation for MT\n- Low Resource ASR\n- Active Learning\n- ...\n\n🔗 [Link to 2020 Course](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PL8PYTP1V4I8CHhppU6n1Q9-04m96D9gt5)\n\n🔗 [Link to 2022 Course](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PL8PYTP1V4I8BhCpzfdKKdd1OnTfLcyZr7)\n\n## Advanced NLP\n\nTo learn advanced concepts in NLP:\n\n- Attention Mechanisms\n- Transformers\n- BERT\n- Question Answering\n- Model Distillation\n- Vision + Language\n- Ethics in NLP\n- Commonsense Reasoning\n- ...\n\n🔗 [Link to Course](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLWnsVgP6CzadmQX6qevbar3_vDBioWHJL)\n\n## CS231N: Convolutional Neural Networks for Visual Recognition\n\nStanford's Famous CS231n course. The videos are only available for the Spring 2017 semester. The course is currently known as Deep Learning for Computer Vision, but the Spring 2017 version is titled Convolutional Neural Networks for Visual Recognition.\n\n- Image Classification\n- Loss Functions and Optimization\n- Introduction to Neural Networks\n- Convolutional Neural Networks\n- Training Neural Networks\n- Deep Learning Software\n- CNN Architectures\n- Recurrent Neural Networks\n- Detection and Segmentation\n- Visualizing and Understanding\n- Generative Models\n- Deep Reinforcement Learning\n\n🔗 [Link to Course](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PL3FW7Lu3i5JvHM8ljYj-zLfQRF3EO8sYv) 🔗 [Link to Materials](http:\u002F\u002Fcs231n.stanford.edu\u002F2017\u002F)\n\n## Deep Learning for Computer Vision\n\nTo learn some of the fundamental concepts in CV:\n\n- Introduction to deep learning for CV\n- Image Classification\n- Convolutional Networks\n- Attention Networks\n- Detection and Segmentation\n- Generative Models\n\n🔗 [Link to Course](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PL5-TkQAfAZFbzxjBHtzdVCWE0Zbhomg7r)\n\n## Deep Learning for Computer Vision (DL4CV)\n\nTo learn modern methods for computer vision:\n\n- CNNs\n- Advanced PyTorch\n- Understanding Neural Networks\n- RNN, Attention and ViTs\n- Generative Models\n- GPU Fundamentals\n- Self-Supervision\n- Neural Rendering\n- Efficient Architectures\n\n🔗 [Link to Course](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PL_Z2_U9MIJdNgFM7-f2fZ9ZxjVRP_jhJv)\n\n## Deep Learning for Computer Vision (neuralearn.ai)\n\nTo learn modern methods for computer vision:\n\n- Self-Supervised Learning\n- Neural Rendering\n- Efficient Architectures\n- Machine Learning Operations (MLOps)\n- Modern Convolutional Neural Networks\n- Transformers in Vision\n- Model Deployment\n\n🔗 [Link to Course](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=IA3WxTTPXqQ)\n\n## AMMI Geometric Deep Learning Course\n\nTo learn about concepts in geometric deep learning:\n\n- Learning in High Dimensions\n- Geometric Priors\n- Grids\n- Manifolds and Meshes\n- Sequences and Time Warping\n- ...\n\n🔗 [Link to Course](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLn2-dEmQeTfSLXW8yXP4q_Ii58wFdxb3C)\n\n## Deep Reinforcement Learning\n\nTo learn the latest concepts in deep RL:\n\n- Intro to RL\n- RL algorithms\n- Real-world sequential decision making\n- Supervised learning of behaviors\n- Deep imitation learning\n- Cost functions and reward functions\n- ...\n\n🔗 [Link to Course](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PL_iWQOsE6TfURIIhCrlt-wj9ByIVpbfGc)\n\n## Reinforcement Learning Lecture Series (DeepMind)\n\nThe Deep Learning Lecture Series is a collaboration between DeepMind and the UCL Centre for Artificial Intelligence.\n\n- Introduction to RL\n- Dynamic Programming\n- Model-free algorithms\n- Deep reinforcement learning\n- ...\n\n🔗 [Link to Course](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLqYmG7hTraZDVH599EItlEWsUOsJbAodm)\n\n\n## LLMOps: Building Real-World Applications With Large Language Models\n\nLearn to build modern software with LLMs using the newest tools and techniques in the field.\n\n🔗 [Link to Course](https:\u002F\u002Fwww.comet.com\u002Fsite\u002Fllm-course\u002F)\n\n## Evaluating and Debugging Generative AI\n\nYou'll learn:\n\n- Instrument A Jupyter Notebook\n- Manage Hyperparameters Config\n- Log Run Metrics\n- Collect artifacts for dataset and model versioning\n- Log experiment results\n- Trace prompts and responses for LLMs\n- ...\n\n🔗 [Link to Course](https:\u002F\u002Fwww.deeplearning.ai\u002Fshort-courses\u002Fevaluating-debugging-generative-ai\u002F)\n\n## ChatGPT Prompt Engineering for Developers\n\nLearn how to use a large language model (LLM) to quickly build new and powerful applications.\n\n🔗 [Link to Course](https:\u002F\u002Fwww.deeplearning.ai\u002Fshort-courses\u002Fchatgpt-prompt-engineering-for-developers\u002F)\n\n## LangChain for LLM Application Development\n\nYou'll learn:\n\n- Models, Prompt, and Parsers\n- Memories for LLMs\n- Chains\n- Question Answering over Documents\n- Agents\n\n🔗 [Link to Course](https:\u002F\u002Fwww.deeplearning.ai\u002Fshort-courses\u002Flangchain-for-llm-application-development\u002F)\n\n## LangChain: Chat with Your Data\n\nYou'll learn about:\n\n- Document Loading\n- Document Splitting\n- Vector Stores and Embeddings\n- Retrieval\n- Question Answering\n- Chat\n\n🔗 [Link to Course](https:\u002F\u002Fwww.deeplearning.ai\u002Fshort-courses\u002Fbuilding-systems-with-chatgpt\u002F)\n\n## Building Systems with the ChatGPT API\n\nLearn how to automate complex workflows using chain calls to a large language model.\n\n🔗 [Link to Course](https:\u002F\u002Fwww.deeplearning.ai\u002Fshort-courses\u002Fbuilding-systems-with-chatgpt\u002F)\n\n## LangChain & Vector Databases in Production\n\nLearn how to use LangChain and Vector DBs in Production:\n\n- LLMs and LangChain\n- Learning how to Prompt\n- Keeping Knowledge Organized with Indexes\n- Combining Components Together with Chains\n- ...\n\n🔗 [Link to Course](https:\u002F\u002Flearn.activeloop.ai\u002Fcourses\u002Flangchain)\n\n## Building LLM-Powered Apps\n\nLearn how to build LLM-powered applications using LLM APIs\n\n- Unpacking LLM APIs\n- Building a Baseline LLM Application\n- Enhancing and Optimizing LLM Applications\n- ...\n\n🔗 [Link to Course](https:\u002F\u002Fwww.wandb.courses\u002Fcourses\u002Fbuilding-llm-powered-apps)\n\n## Full Stack LLM Bootcamp\n\nTo learn how to build and deploy LLM-powered applications:\n\n- Learn to Spell: Prompt Engineering\n- LLMOPs\n- UX for Language User Interfaces\n- Augmented Language Models\n- Launch an LLM App in One Hour\n- LLM Foundations\n- Project Walkthrough: askFSDL\n- ...\n\n🔗 [Link to Course](https:\u002F\u002Ffullstackdeeplearning.com\u002Fllm-bootcamp\u002Fspring-2023\u002F)\n\n## Full Stack Deep Learning\n\nTo learn full-stack production deep learning:\n\n- ML Projects\n- Infrastructure and Tooling\n- Experiment Managing\n- Troubleshooting DNNs\n- Data Management\n- Data Labeling\n- Monitoring ML Models\n- Web deployment\n- ...\n\n🔗 [Link to Course](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PL1T8fO7ArWlcWg04OgNiJy91PywMKT2lv)\n\n## Introduction to Deep Learning and Deep Generative Models\n\nCovers the fundamental concepts of deep learning\n\n- Single-layer neural networks and gradient descent\n- Multi-layer neural networks and backpropagation\n- Convolutional neural networks for images\n- Recurrent neural networks for text\n- Autoencoders, variational autoencoders, and generative adversarial networks\n- Encoder-decoder recurrent neural networks and transformers\n- PyTorch code examples\n\n🔗 [Link to Course](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=1nqCZqDYPp0&list=PLTKMiZHVd_2KJtIXOW0zFhFfBaJJilH51) 🔗 [Link to Materials](https:\u002F\u002Fsebastianraschka.com\u002Fblog\u002F2021\u002Fdl-course.html)\n\n## Self-Driving Cars (Tübingen)\n\nCovers the most dominant paradigms of self-driving cars: modular pipeline-based approaches as well as deep-learning based end-to-end driving techniques.\n\n- Camera, lidar and radar-based perception\n- Localization, navigation, path planning\n- Vehicle modeling\u002Fcontrol\n- Deep Learning\n- Imitation learning\n- Reinforcement learning\n\n🔗 [Link to Course](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PL05umP7R6ij321zzKXK6XCQXAaaYjQbzr)\n\n## Reinforcement Learning (Polytechnique Montreal, Fall 2021)\n\nDesigning autonomous decision making systems is one of the longstanding goals of Artificial Intelligence. Such decision making systems, if realized, can have a big impact in machine learning for robotics, game playing, control, health care to name a few. This course introduces Reinforcement Learning as a general framework to design such autonomous decision making systems.\n\n- Introduction to RL\n- Multi-armed bandits\n- Policy Gradient Methods\n- Contextual Bandits\n- Finite Markov Decision Process\n- Dynamic Programming\n- Policy Iteration, Value Iteration\n- Monte Carlo Methods\n- ...\n\n🔗 [Link to Course](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLImtCgowF_ES_JdF_UcM60EXTcGZg67Ua) 🔗 [Link to Materials](https:\u002F\u002Fchandar-lab.github.io\u002FINF8953DE\u002F)\n\n## Foundations of Deep RL\n\nA mini 6-lecture series by Pieter Abbeel.\n\n- MDPs, Exact Solution Methods, Max-ent RL\n- Deep Q-Learning\n- Policy Gradients and Advantage Estimation\n- TRPO and PPO\n- DDPG and SAC\n- Model-based RL\n\n🔗 [Link to Course](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLwRJQ4m4UJjNymuBM9RdmB3Z9N5-0IlY0)\n\n## Stanford CS234: Reinforcement Learning\n\nCovers topics from basic concepts of Reinforcement Learning to more advanced ones:\n\n- Markov decision processes & planning\n- Model-free policy evaluation\n- Model-free control\n- Reinforcement learning with function approximation & Deep RL\n- Policy Search\n- Exploration\n- ...\n\n🔗 [Link to Course](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLoROMvodv4rOSOPzutgyCTapiGlY2Nd8u) 🔗 [Link to Materials](https:\u002F\u002Fweb.stanford.edu\u002Fclass\u002Fcs234\u002F)\n\n## Stanford CS330: Deep Multi-Task and Meta Learning\n\nThis is a graduate-level course covering different aspects of deep multi-task and meta learning.\n\n- Multi-task learning, transfer learning basics\n- Meta-learning algorithms\n- Advanced meta-learning topics\n- Multi-task RL, goal-conditioned RL\n- Meta-reinforcement learning\n- Hierarchical RL\n- Lifelong learning\n- Open problems\n\n🔗 [Link to Course](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLoROMvodv4rMC6zfYmnD7UG3LVvwaITY5) 🔗 [Link to Materials](https:\u002F\u002Fcs330.stanford.edu\u002F)\n\n## MIT Deep Learning in Life Sciences\n\nA course introducing foundations of ML for applications in genomics and the life sciences more broadly.\n\n- Interpreting ML Models\n- DNA Accessibility, Promoters and Enhancers\n- Chromatin and gene regulation\n- Gene Expression, Splicing\n- RNA-seq, Splicing\n- Single cell RNA-sequencing\n- Dimensionality Reduction, Genetics, and Variation\n- Drug Discovery\n- Protein Structure Prediction\n- Protein Folding\n- Imaging and Cancer\n- Neuroscience\n\n🔗 [Link to Course](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLypiXJdtIca5ElZMWHl4HMeyle2AzUgVB)\n\n🔗 [Link to Materials](https:\u002F\u002Fmit6874.github.io\u002F)\n\n## Advanced Robotics: UC Berkeley\n\nThis is course is from Peter Abbeel and covers a review on reinforcement learning and continues to applications in robotics.\n\n- MDPs: Exact Methods\n- Discretization of Continuous State Space MDPs\n- Function Approximation \u002F Feature-based Representations\n- LQR, iterative LQR \u002F Differential Dynamic Programming\n- ...\n\n🔗 [Link to Course](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLwRJQ4m4UJjNBPJdt8WamRAt4XKc639wF) 🔗 [Link to Materials](https:\u002F\u002Fpeople.eecs.berkeley.edu\u002F~pabbeel\u002Fcs287-fa19\u002F)\n\n---\n\nReach out on [Twitter](https:\u002F\u002Ftwitter.com\u002Fomarsar0) if you have any questions.\n\nIf you are interested to contribute, feel free to open a PR with a link to the course. It will take a bit of time, but I have plans to do many things with these individual lectures. We can summarize the lectures, include notes, provide additional reading material, include difficulty of content, etc.\n\nYou can now find ML Course notes [here](https:\u002F\u002Fgithub.com\u002Fdair-ai\u002FML-Course-Notes).\n","# 📺 机器学习 YouTube 课程\n\n在 DAIR.AI，我们热衷于开源人工智能教育。在这个仓库中，我们整理并归类了 YouTube 上一些最佳且最新的机器学习课程。\n\n**机器学习**\n\n- [加州理工学院 CS156：从数据中学习](#caltech-cs156-learning-from-data)\n- [斯坦福大学 CS229：机器学习](#stanford-cs229-machine-learning)\n- [与机器学习交朋友](#making-friends-with-machine-learning)\n- [应用机器学习](#applied-machine-learning)\n- [图宾根大学机器学习导论](#introduction-to-machine-learning-Tübingen)\n- [斯蒂芬·哈梅林的机器学习讲座](#machine-learning-lecture-stefan-harmeling)\n- [图宾根大学统计机器学习](#statistical-machine-learning-Tübingen)\n- [概率机器学习](#probabilistic-machine-learning)\n- [麻省理工学院 6.S897：医疗领域的机器学习（2019）](#mit-6s897-machine-learning-for-healthcare-2019)\n\n**深度学习**\n\n- [神经网络：从零到英雄（安德烈·卡帕西）](#neural-networks-zero-to-hero-by-andrej-karpathy)\n- [麻省理工学院：艺术、美学与创造力中的深度学习](#mit-deep-learning-for-art-aesthetics-and-creativity)\n- [斯坦福大学 CS230：深度学习（2018）](#stanford-cs230-deep-learning-2018)\n- [麻省理工学院深度学习导论](#introduction-to-deep-learning)\n- [卡内基梅隆大学深度学习导论（11-785）](#cmu-introduction-to-deep-learning-11-785)\n- [深度学习：CS 182](#deep-learning-cs-182)\n- [深度无监督学习](#deep-unsupervised-learning)\n- [纽约大学深度学习 SP21](#nyu-deep-learning-sp21)\n- [基础模型](#foundation-models)\n- [图宾根大学深度学习](#deep-learning-Tübingen)\n\n**科学机器学习**\n\n- [并行计算与科学机器学习](#parallel-computing-and-scientific-machine-learning)\n\n**实践机器学习**\n\n- [LLMOps：使用大型语言模型构建真实世界的应用程序](#llmops-building-real-world-applications-with-large-language-models)\n- [生成式 AI 的评估与调试](#evaluating-and-debugging-generative-ai)\n- [面向开发者的 ChatGPT 提示工程](#chatgpt-prompt-engineering-for-developers)\n- [LangChain 用于 LLM 应用开发](#langchain-for-llm-application-development)\n- [LangChain：与你的数据对话](#langchain-chat-with-your-data)\n- [使用 ChatGPT API 构建系统](#building-systems-with-the-chatgpt-api)\n- [生产环境中的 LangChain 和向量数据库](#langchain--vector-databases-in-production)\n- [构建 LLM 驱动的应用程序](#building-llm-powered-apps)\n- [全栈 LLM 训练营](#full-stack-llm-bootcamp)\n- [全栈深度学习](#full-stack-deep-learning)\n- [面向编码人员的实用深度学习](#practical-deep-learning-for-coders)\n- [斯坦福大学 MLSys 研讨会](#stanford-mlsys-seminars)\n- [面向生产的机器学习工程（MLOps）](#machine-learning-engineering-for-production-mlops)\n- [麻省理工学院以数据为中心的 AI 导论](#mit-introduction-to-data-centric-ai)\n\n**自然语言处理**\n\n- [XCS224U：自然语言理解（2023）](#xcs224u-natural-language-understanding-2023)\n- [斯坦福大学 CS25 - 变压器联盟](#stanford-cs25---transformers-united)\n- [Hugging Face 自然语言处理课程](#nlp-course-hugging-face)\n- [CS224N：结合深度学习的自然语言处理](#cs224n-natural-language-processing-with-deep-learning)\n- [卡内基梅隆大学用于 NLP 的神经网络](#cmu-neural-networks-for-nlp)\n- [CS224U：自然语言理解](#cs224u-natural-language-understanding)\n- [卡内基梅隆大学高级 NLP 2021\u002F2022\u002F2024](#cmu-advanced-nlp)\n- [多语言 NLP](#multilingual-nlp)\n- [高级 NLP](#advanced-nlp)\n\n**计算机视觉**\n\n- [CS231N：用于视觉识别的卷积神经网络](#cs231n-convolutional-neural-networks-for-visual-recognition)\n- [计算机视觉中的深度学习](#deep-learning-for-computer-vision)\n- [计算机视觉中的深度学习（DL4CV）](#deep-learning-for-computer-vision-dl4cv)\n- [计算机视觉中的深度学习（neuralearn.ai）](#deep-learning-for-computer-vision-neuralearnai)\n\n**强化学习**\n\n- [深度强化学习](#deep-reinforcement-learning)\n- [DeepMind 强化学习系列讲座](#reinforcement-learning-lecture-series-deepmind)\n- [蒙特利尔理工大学 2021 年秋季的强化学习](#reinforcement-learning-polytechnique-montreal-fall-2021)\n- [深度 RL 的基础](#foundations-of-deep-rl)\n- [斯坦福大学 CS234：强化学习](#stanford-cs234-reinforcement-learning)\n\n**图机器学习**\n\n- [斯坦福大学图上的机器学习](#machine-learning-with-graphs-stanford)\n- [AMMI 几何深度学习课程](#ammi-geometric-deep-learning-course)\n\n**多任务学习**\n\n- [斯坦福大学 CS330：深度多任务与元学习](#stanford-cs330-deep-multi-task-and-meta-learning)\n\n**其他**\n\n- [麻省理工学院生命科学中的深度学习](#mit-deep-learning-in-life-sciences)\n- [自动驾驶汽车（图宾根）](#self-driving-cars-Tübingen)\n- [伯克利大学高级机器人技术](#advanced-robotics-uc-berkeley)\n\n---\n\n## 加州理工学院 CS156：从数据中学习\n\n这是一门机器学习入门课程，涵盖了基本理论、算法和应用。\n\n- 第1讲：学习问题\n- 第2讲：学习是否可行？\n- 第3讲：线性模型 I\n- 第4讲：误差与噪声\n- 第5讲：训练与测试\n- 第6讲：泛化理论\n- 第7讲：VC 维度\n- 第8讲：偏差-方差权衡\n- 第9讲：线性模型 II\n- 第10讲：神经网络\n- 第11讲：过拟合\n- 第12讲：正则化\n- 第13讲：验证\n- 第14讲：支持向量机\n- 第15讲：核方法\n- 第16讲：径向基函数\n- 第17讲：三个学习原则\n- 第18讲：尾声\n\n🔗 [课程链接](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLD63A284B7615313A)\n\n## 斯坦福大学 CS229：机器学习\n\n学习一些 ML 的基础知识：\n\n- 线性回归与梯度下降\n- 逻辑回归\n- 朴素贝叶斯\n- SVM\n- 核函数\n- 决策树\n- 神经网络导论\n- 调试 ML 模型\n  ...\n\n🔗 [课程链接](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLoROMvodv4rMiGQp3WXShtMGgzqpfVfbU)\n\n## 与机器学习交朋友\n\n一系列迷你讲座，涵盖 ML 的各种入门主题：\n\n- AI 的可解释性\n- 分类与回归的区别\n- 精确率与召回率\n- 统计显著性\n- 聚类与 K-means\n- 集成模型\n  ...\n\n🔗 [课程链接](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLRKtJ4IpxJpDxl0NTvNYQWKCYzHNuy2xG)\n\n## 神经网络：从零到英雄（由安德烈·卡帕西主讲）\n\n本课程深入介绍了神经网络。\n\n- 反向传播\n- 详细讲解语言建模\n- 激活函数与梯度\n- 成为反向传播高手\n\n🔗 [课程链接](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLAqhIrjkxbuWI23v9cThsA9GvCAUhRvKZ)\n\n## MIT：艺术、美学与创造力中的深度学习\n\n涵盖深度学习在艺术、美学和创造力领域的应用。\n\n- 怀旧 -> 艺术 -> 创造力 -> 演化作为数据 + 方向\n- 高效生成对抗网络\n- 人工智能在创造力方面的探索\n- 神经抽象\n- 使用一致神经场轻松创建3D内容\n  ...\n\n🔗 [课程链接](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLCpMvp7ftsnIbNwRnQJbDNRqO6qiN3EyH)\n\n## 斯坦福CS230：深度学习（2018年）\n\n涵盖深度学习的基础知识、如何构建不同类型的神经网络（CNN、RNN、LSTM等）、如何领导机器学习项目，以及针对深度学习从业者的职业建议。\n\n- 渗透深度学习的直觉\n- 对抗样本 - GANs\n- 深度学习项目的完整流程\n- 人工智能与医疗健康\n- 大模型战略\n- 神经网络的可解释性\n- 职业建议与阅读研究论文\n- 深度强化学习\n\n🔗 [课程链接](https:\u002F\u002Fyoutube.com\u002Fplaylist?list=PLoROMvodv4rOABXSygHTsbvUz4G_YQhOb) 🔗 [资料链接](https:\u002F\u002Fcs230.stanford.edu\u002Fsyllabus\u002F)\n\n## 应用机器学习\n\n学习一些最广泛使用的机器学习技术：\n\n- 优化与微积分\n- 过拟合与欠拟合\n- 正则化\n- 蒙特卡洛估计\n- 最大似然学习\n- 最近邻算法\n- …\n\n🔗 [课程链接](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PL2UML_KCiC0UlY7iCQDSiGDMovaupqc83)\n\n## 图宾根大学机器学习导论\n\n本课程是机器学习的基础入门课程，涵盖了回归、分类、优化、正则化、聚类和降维等关键概念。\n\n- 线性回归\n- 逻辑回归\n- 正则化\n- 提升方法\n- 神经网络\n- 主成分分析\n- 聚类\n- …\n\n🔗 [课程链接](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PL05umP7R6ij35ShKLDqccJSDntugY4FQT)\n\n## 斯特凡·哈梅林机器学习讲座\n\n涵盖许多基础的机器学习概念：\n\n- 贝叶斯法则\n- 从逻辑到概率\n- 分布\n- 矩阵微分学\n- PCA\n- K均值与EM算法\n- 因果关系\n- 高斯过程\n- …\n\n🔗 [课程链接](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLzrCXlf6ypbxS5OYOY3EN_0u2fDuIT6Gt)\n\n## 图宾根大学统计机器学习\n\n本课程覆盖了统计机器学习中的标准范式和算法。\n\n- K近邻算法\n- 贝叶斯决策理论\n- 凸优化\n- 线性回归与岭回归\n- 逻辑回归\n- 支持向量机\n- 随机森林\n- 提升方法\n- PCA\n- 聚类\n- …\n\n🔗 [课程链接](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PL05umP7R6ij2XCvrRzLokX6EoHWaGA2cC)\n\n## 针对程序员的实用深度学习\n\n本课程涵盖以下主题：\n\n- 如何为计算机视觉、自然语言处理、表格数据分析和协同过滤问题构建并训练深度学习模型\n- 创建随机森林和回归模型\n- 部署模型\n- 使用全球发展最快的深度学习框架PyTorch，以及fastai和Hugging Face等流行库\n- 扩散模型的基础与深入探讨\n- …\n\n🔗 [课程链接 - 第一部分](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLfYUBJiXbdtSvpQjSnJJ_PmDQB_VyT5iU)\n\n🔗 [课程链接 - 第二部分](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=_7rMfsA24Ls&ab_channel=JeremyHoward)\n\n## 斯坦福MLSys研讨会\n\n一系列关于构建机器学习系统相关主题的研讨会。\n\n🔗 [讲座链接](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLSrTvUm384I9PV10koj_cqit9OfbJXEkq)\n\n## 生产环境下的机器学习工程（MLOps）\n\n吴恩达主讲的MLOps专项课程。\n\n🔗 [讲座链接](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLkDaE6sCZn6GMoA0wbpJLi3t34Gd8l0aK)\n\n## MIT数据为中心的人工智能导论\n\n介绍新兴的数据为中心的人工智能科学（DCAI），该领域研究改进数据集的技术，而这通常是提升实际机器学习应用性能的最佳方式。主题包括：\n\n- 数据为中心AI与模型为中心AI\n- 标签错误\n- 数据集的创建与管理\n- 基于数据的机器学习模型评估\n- 类别不平衡、异常值与分布偏移\n- …\n\n🔗 [课程官网](https:\u002F\u002Fdcai.csail.mit.edu\u002F)\n\n🔗 [讲座视频](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=ayzOzZGHZy4&list=PLnSYPjg2dHQKdig0vVbN-ZnEU0yNJ1mo5)\n\n🔗 [实验作业](https:\u002F\u002Fgithub.com\u002Fdcai-course\u002Fdcai-lab)\n\n## 斯坦福图机器学习\n\n学习机器学习中最新的图技术：\n\n- PageRank\n- 矩阵分解\n- 节点嵌入\n- 图神经网络\n- 知识图谱\n- 针对图的深度生成模型\n- …\n\n🔗 [课程链接](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLoROMvodv4rPLKxIpqhjhPgdQy7imNkDn)\n\n## 概率机器学习\n\n学习机器学习的概率范式：\n\n- 不确定性的推理\n- 连续变量\n- 抽样\n- 马尔可夫链蒙特卡洛\n- 高斯分布\n- 图模型\n- 推理算法的调优\n- …\n\n🔗 [课程链接](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PL05umP7R6ij2YE8rRJSb-olDNbntAQ_Bx)\n\n## MIT 6.S897：2019年医疗健康领域的机器学习\n\n本课程向学生介绍医疗健康领域的机器学习，包括临床数据的特性，以及如何利用机器学习进行风险分层、疾病进展建模、精准医学、诊断、亚型发现和优化临床工作流程。\n\n🔗 [课程链接](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLUl4u3cNGP60B0PQXVQyGNdCyCTDU1Q5j)\n\n## 深度学习导论\n\n学习深度学习的一些基础知识：\n\n- 深度学习简介\n\n🔗 [课程链接](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLtBw6njQRU-rwp5__7C0oIVt26ZgjG9NI)\n\n## 卡内基梅隆大学11-785深度学习导论\n\n本课程从多层感知机（MLP）开始逐步展开，随后进入注意力机制和序列到序列模型等概念。\n\n🔗 [课程链接](https:\u002F\u002Fdeeplearning.cs.cmu.edu\u002FF22\u002Findex.html) \\\n🔗 [讲座](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLp-0K3kfddPxRmjgjm0P1WT6H-gTqE8j9) \\\n🔗 [辅导课\u002F习题课](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLp-0K3kfddPz8WXg8RqH0sEN6X2L65HUZ)\n\n## 深度学习：CS 182\n\n学习深度学习中广泛使用的一些技术：\n\n- 机器学习基础\n- 错误分析\n- 优化\n- 反向传播\n- 参数初始化\n- 批量归一化\n- 风格迁移\n- 模仿学习\n- …\n\n🔗 [课程链接](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PL_iWQOsE6TfVmKkQHucjPAoRtIJYt8a5A)\n\n## 深度无监督学习\n\n学习最新且最常用的深度无监督学习技术：\n\n- 自回归模型\n- 流模型\n- 隐变量模型\n- 自监督学习\n- 隐式模型\n- 压缩\n- …\n\n🔗 [课程链接](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLwRJQ4m4UJjPiJP3691u-qWwPGVKzSlNP)\n\n## NYU 深度学习 SP21\n\n学习深度学习中的一些高级技术：\n\n- 神经网络：旋转与挤压\n- 隐变量能量模型\n- 无监督学习\n- 生成对抗网络\n- 自编码器\n- …\n\n🔗 [课程链接](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLLHTzKZzVU9e6xUfG10TkTWApKSZCzuBI)\n\n## 基础模型\n\n学习 GPT-3、CLIP、Flamingo、Codex 和 DINO 等基础模型。\n\n🔗 [课程链接](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PL9t0xVFP90GD8hox0KipBkJcLX_C3ja67)\n\n## 深度学习（图宾根）\n\n本课程介绍深度神经网络的实践与理论原理。\n\n- 计算图\n- 激活函数与损失函数\n- 训练、正则化与数据增强\n- 基础及最先进的深度神经网络架构，包括卷积网络和图神经网络\n- 深度生成模型，如自编码器、变分自编码器和生成对抗网络\n- …\n\n🔗 [课程链接](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PL05umP7R6ij3NTWIdtMbfvX7Z-4WEXRqD)\n\n## 并行计算与科学机器学习\n\n- 科学模拟器基础\n- 并行计算入门\n- 连续动力学\n- 反问题与可微编程\n- 分布式并行计算\n- 物理信息神经网络与神经微分方程\n- 概率编程，又称程序上的贝叶斯估计\n- 全局化对模型的理解\n\n🔗 [课程链接](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLCAl7tjCwWyGjdzOOnlbGnVNZk0kB8VSa)\n\n## XCS224U：自然语言理解（2023年）\n\n本课程涵盖以下主题：\n\n- 上下文词表示\n- 信息检索\n- 上下文学习\n- NLU 模型的行为评估\n- NLP 方法与指标\n- …\n\n🔗 [课程链接](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLoROMvodv4rOwvldxftJTmoR3kRcWkJBp)\n\n## 斯坦福 CS25 - Transformer 联盟\n\n本课程专注于 Transformer 的讲座，深入探讨其原理及应用。\n\n- Transformer 入门\n- 语言中的 Transformer：GPT-3、Codex\n- 视觉领域的应用\n- RL 中的 Transformer 及通用计算引擎\n- Transformer 的规模扩展\n- Transformer 的可解释性\n- …\n\n🔗 [课程链接](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLoROMvodv4rNiJRchCzutFw5ItR_Z27CM)\n\n## Hugging Face NLP 课程\n\n学习不同的 NLP 概念，以及如何将语言模型和 Transformer 应用于 NLP：\n\n- 什么是迁移学习？\n- BPE 分词\n- 批量输入处理\n- 模型微调\n- 文本嵌入与语义搜索\n- 模型评估\n- …\n\n🔗 [课程链接](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLo2EIpI_JMQvWfQndUesu0nPBAtZ9gP1o)\n\n## CS224N：基于深度学习的自然语言处理\n\n学习最新的基于深度学习的 NLP 方法：\n\n- 依存句法分析\n- 语言模型与 RNN\n- 问答系统\n- Transformer 与预训练\n- 自然语言生成\n- T5 与大型语言模型\n- NLP 的未来\n- …\n\n🔗 [课程链接](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLoROMvodv4rOSH4v6133s9LFPRHjEmbmJ)\n\n## CMU NLP 中的神经网络\n\n学习最新的基于神经网络的 NLP 技术：\n\n- 语言建模\n- 效率技巧\n- 条件生成\n- 结构化预测\n- 模型解释\n- 高级搜索算法\n- …\n\n🔗 [课程链接](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PL8PYTP1V4I8AkaHEJ7lOOrlex-pcxS-XV)\n\n## CS224U：自然语言理解\n\n学习自然语言理解领域的最新概念：\n\n- 基于场景的语言理解\n- 关系抽取\n- 自然语言推理 (NLI)\n- NLU 与神经信息抽取\n- 对抗性测试\n- …\n\n🔗 [课程链接](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLoROMvodv4rPt5D0zs3YhbWSZA8Q_DyiJ)\n\n## CMU 高级 NLP\n\n学习内容包括：\n\n- 现代 NLP 技术的基础\n- 多任务、多领域、多语言学习\n- 提示工程 + 序列到序列预训练\n- NLP 模型的解释与调试\n- 从知识库中学习\n- 对抗性学习\n- …\n\n🔗 [2021 年版链接](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PL8PYTP1V4I8AYSXn_GKVgwXVluCT9chJ6)\n\n🔗 [2022 年版链接](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PL8PYTP1V4I8D0UkqW2fEhgLrnlDW9QK7z)\n\n🔗 [2024 年版链接](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PL8PYTP1V4I8DZprnWryM4nR8IZl1ZXDjg)\n\n## 多语言 NLP\n\n学习进行多语言 NLP 的最新概念：\n\n- 类型学\n- 词汇、词性与形态学\n- 高级文本分类\n- 机器翻译\n- MT 的数据增强\n- 低资源 ASR\n- 主动学习\n- …\n\n🔗 [2020 年课程链接](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PL8PYTP1V4I8CHhppU6n1Q9-04m96D9gt5)\n\n🔗 [2022 年课程链接](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PL8PYTP1V4I8BhCpzfdKKdd1OnTfLcyZr7)\n\n## 高级 NLP\n\n学习 NLP 中的高级概念：\n\n- 注意力机制\n- Transformer\n- BERT\n- 问答系统\n- 模型蒸馏\n- 视觉与语言结合\n- NLP 中的伦理问题\n- 常识推理\n- …\n\n🔗 [课程链接](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLWnsVgP6CzadmQX6qevbar3_vDBioWHJL)\n\n## CS231N：用于视觉识别的卷积神经网络\n\n斯坦福著名的 CS231n 课程。视频仅提供 2017 年春季学期的内容。该课程目前被称为“计算机视觉中的深度学习”，但 2017 年春季版本的标题是“用于视觉识别的卷积神经网络”。\n\n- 图像分类\n- 损失函数与优化\n- 神经网络入门\n- 卷积神经网络\n- 神经网络的训练\n- 深度学习软件\n- CNN 架构\n- 循环神经网络\n- 目标检测与分割\n- 可视化与理解\n- 生成模型\n- 深度强化学习\n\n🔗 [课程链接](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PL3FW7Lu3i5JvHM8ljYj-zLfQRF3EO8sYv) 🔗 [资料链接](http:\u002F\u002Fcs231n.stanford.edu\u002F2017\u002F)\n\n## 计算机视觉中的深度学习\n\n学习 CV 中的一些基本概念：\n\n- 计算机视觉中的深度学习简介\n- 图像分类\n- 卷积网络\n- 注意力网络\n- 目标检测与分割\n- 生成模型\n\n🔗 [课程链接](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PL5-TkQAfAZFbzxjBHtzdVCWE0Zbhomg7r)\n\n## 计算机视觉中的深度学习（DL4CV）\n\n学习现代计算机视觉方法：\n\n- CNN\n- 高级 PyTorch\n- 神经网络的理解\n- RNN、注意力机制与 ViT\n- 生成模型\n- GPU 基础\n- 自监督学习\n- 神经渲染\n- 高效架构\n\n🔗 [课程链接](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PL_Z2_U9MIJdNgFM7-f2fZ9ZxjVRP_jhJv)\n\n## 计算机视觉深度学习（neuralearn.ai）\n\n学习计算机视觉的现代方法：\n\n- 自监督学习\n- 神经渲染\n- 高效架构\n- 机器学习运维（MLOps）\n- 现代卷积神经网络\n- 视觉中的Transformer\n- 模型部署\n\n🔗 [课程链接](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=IA3WxTTPXqQ)\n\n## AMMI 几何深度学习课程\n\n学习几何深度学习的相关概念：\n\n- 高维空间中的学习\n- 几何先验\n- 网格\n- 流形与网格模型\n- 序列与时间规整\n- …\n\n🔗 [课程链接](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLn2-dEmQeTfSLXW8yXP4q_Ii58wFdxb3C)\n\n## 深度强化学习\n\n学习深度强化学习的最新概念：\n\n- 强化学习入门\n- 强化学习算法\n- 现实世界中的序列决策\n- 行为的监督学习\n- 深度模仿学习\n- 成本函数与奖励函数\n- …\n\n🔗 [课程链接](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PL_iWQOsE6TfURIIhCrlt-wj9ByIVpbfGc)\n\n## 强化学习讲座系列（DeepMind）\n\n深度学习讲座系列是由DeepMind与UCL人工智能中心合作推出的。\n\n- 强化学习导论\n- 动态规划\n- 无模型算法\n- 深度强化学习\n- …\n\n🔗 [课程链接](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLqYmG7hTraZDVH599EItlEWsUOsJbAodm)\n\n\n## LLMOps：使用大型语言模型构建真实世界应用\n\n学习如何利用该领域的最新工具和技术，使用LLM构建现代软件。\n\n🔗 [课程链接](https:\u002F\u002Fwww.comet.com\u002Fsite\u002Fllm-course\u002F)\n\n## 生成式AI的评估与调试\n\n你将学习：\n\n- 在Jupyter Notebook中进行仪器化\n- 管理超参数配置\n- 记录运行指标\n- 收集数据集和模型版本化的相关工件\n- 记录实验结果\n- 跟踪LLM的提示与响应\n- …\n\n🔗 [课程链接](https:\u002F\u002Fwww.deeplearning.ai\u002Fshort-courses\u002Fevaluating-debugging-generative-ai\u002F)\n\n## ChatGPT 提示工程：面向开发者\n学习如何利用大型语言模型（LLM）快速构建全新且强大的应用。\n\n🔗 [课程链接](https:\u002F\u002Fwww.deeplearning.ai\u002Fshort-courses\u002Fchatgpt-prompt-engineering-for-developers\u002F)\n\n## LangChain用于LLM应用开发\n你将学习：\n\n- 模型、提示与解析器\n- LLM的记忆机制\n- 链式结构\n- 文档问答系统\n- 代理\n\n🔗 [课程链接](https:\u002F\u002Fwww.deeplearning.ai\u002Fshort-courses\u002Flangchain-for-llm-application-development\u002F)\n\n## LangChain：与你的数据聊天\n你将了解：\n\n- 文档加载\n- 文档分割\n- 向量数据库与嵌入\n- 检索\n- 问答\n- 对话\n\n🔗 [课程链接](https:\u002F\u002Fwww.deeplearning.ai\u002Fshort-courses\u002Fbuilding-systems-with-chatgpt\u002F)\n\n## 使用ChatGPT API构建系统\n学习如何通过调用大型语言模型的链式接口来自动化复杂的工作流程。\n\n🔗 [课程链接](https:\u002F\u002Fwww.deeplearning.ai\u002Fshort-courses\u002Fbuilding-systems-with-chatgpt\u002F)\n\n## LangChain与向量数据库在生产环境中的应用\n学习如何在生产环境中使用LangChain和向量数据库：\n\n- LLM与LangChain\n- 提示工程技巧\n- 利用索引组织知识\n- 通过链式结构组合各个组件\n- …\n\n🔗 [课程链接](https:\u002F\u002Flearn.activeloop.ai\u002Fcourses\u002Flangchain)\n\n## 构建LLM驱动的应用程序\n学习如何使用LLM API构建LLM驱动的应用程序：\n\n- 深入理解LLM API\n- 构建基础LLM应用\n- 增强与优化LLM应用\n- …\n\n🔗 [课程链接](https:\u002F\u002Fwww.wandb.courses\u002Fcourses\u002Fbuilding-llm-powered-apps)\n\n## 全栈LLM训练营\n学习如何构建并部署LLM驱动的应用程序：\n\n- 掌握提示工程\n- LLM运维\n- 语言用户界面的用户体验设计\n- 增强型语言模型\n- 一小时内上线LLM应用\n- LLM基础知识\n- 项目实战：askFSDL\n- …\n\n🔗 [课程链接](https:\u002F\u002Ffullstackdeeplearning.com\u002Fllm-bootcamp\u002Fspring-2023\u002F)\n\n## 全栈深度学习\n学习全栈生产级深度学习：\n\n- ML项目\n- 基础设施与工具\n- 实验管理\n- DNN故障排除\n- 数据管理\n- 数据标注\n- ML模型监控\n- Web部署\n- …\n\n🔗 [课程链接](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PL1T8fO7ArWlcWg04OgNiJy91PywMKT2lv)\n\n## 深度学习与深度生成模型导论\n涵盖深度学习的基础概念：\n\n- 单层神经网络与梯度下降\n- 多层神经网络与反向传播\n- 用于图像处理的卷积神经网络\n- 用于文本处理的循环神经网络\n- 自编码器、变分自编码器和生成对抗网络\n- 编码器-解码器循环神经网络与Transformer\n- PyTorch代码示例\n\n🔗 [课程链接](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=1nqCZqDYPp0&list=PLTKMiZHVd_2KJtIXOW0zFhFfBaJJilH51) 🔗 [资料链接](https:\u002F\u002Fsebastianraschka.com\u002Fblog\u002F2021\u002Fdl-course.html)\n\n## 自动驾驶汽车（图宾根）\n涵盖自动驾驶汽车中最主流的范式：基于模块化流水线的方法以及基于深度学习的端到端驾驶技术。\n\n- 基于摄像头、激光雷达和雷达的感知\n- 定位、导航与路径规划\n- 车辆建模与控制\n- 深度学习\n- 模仿学习\n- 强化学习\n\n🔗 [课程链接](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PL05umP7R6ij321zzKXK6XCQXAaaYjQbzr)\n\n## 强化学习（蒙特利尔理工大学，2021年秋季）\n设计自主决策系统是人工智能长期以来的目标之一。如果这些系统得以实现，将在机器人技术、游戏、控制、医疗保健等领域产生重大影响。本课程将强化学习介绍为一种通用框架，用于设计此类自主决策系统。\n\n- 强化学习导论\n- 多臂老虎机问题\n- 策略梯度方法\n- 上下文老虎机问题\n- 有限马尔可夫决策过程\n- 动态规划\n- 策略迭代、值迭代\n- 蒙特卡洛方法\n- …\n\n🔗 [课程链接](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLImtCgowF_ES_JdF_UcM60EXTcGZg67Ua) 🔗 [资料链接](https:\u002F\u002Fchandar-lab.github.io\u002FINF8953DE\u002F)\n\n## 深度强化学习基础\n由Pieter Abbeel主讲的迷你六讲系列。\n\n- MDPs、精确解法、最大熵强化学习\n- 深度Q学习\n- 策略梯度与优势估计\n- TRPO与PPO\n- DDPG与SAC\n- 基于模型的强化学习\n\n🔗 [课程链接](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLwRJQ4m4UJjNymuBM9RdmB3Z9N5-0IlY0)\n\n## 斯坦福CS234：强化学习\n涵盖从强化学习基础概念到更高级主题的内容：\n\n- 马尔可夫决策过程与规划\n- 无模型策略评估\n- 无模型控制\n- 带有函数逼近的强化学习与深度强化学习\n- 策略搜索\n- 探索\n- …\n\n🔗 [课程链接](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLoROMvodv4rOSOPzutgyCTapiGlY2Nd8u) 🔗 [资料链接](https:\u002F\u002Fweb.stanford.edu\u002Fclass\u002Fcs234\u002F)\n\n## 斯坦福CS330：深度多任务学习与元学习\n\n这是一门研究生级别的课程，涵盖了深度多任务学习和元学习的各个方面。\n\n- 多任务学习、迁移学习基础\n- 元学习算法\n- 元学习高级主题\n- 多任务强化学习、目标条件强化学习\n- 元强化学习\n- 层次化强化学习\n- 终身学习\n- 未解决的问题\n\n🔗 [课程链接](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLoROMvodv4rMC6zfYmnD7UG3LVvwaITY5) 🔗 [资料链接](https:\u002F\u002Fcs330.stanford.edu\u002F)\n\n## MIT生命科学中的深度学习\n\n本课程介绍机器学习的基础知识，并探讨其在基因组学及更广泛的生命科学领域的应用。\n\n- 解释机器学习模型\n- DNA可及性、启动子与增强子\n- 染色质与基因调控\n- 基因表达、剪接\n- RNA测序、剪接\n- 单细胞RNA测序\n- 降维、遗传学与变异\n- 药物发现\n- 蛋白质结构预测\n- 蛋白质折叠\n- 影像与癌症\n- 神经科学\n\n🔗 [课程链接](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLypiXJdtIca5ElZMWHl4HMeyle2AzUgVB)\n\n🔗 [资料链接](https:\u002F\u002Fmit6874.github.io\u002F)\n\n## 高级机器人学：加州大学伯克利分校\n\n本课程由Peter Abbeel主讲，首先回顾强化学习，随后深入探讨其在机器人领域的应用。\n\n- 马尔可夫决策过程：精确方法\n- 连续状态空间MDP的离散化\n- 函数近似\u002F基于特征的表示\n- LQR、迭代LQR\u002F微分动态规划\n- …\n\n🔗 [课程链接](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLwRJQ4m4UJjNBPJdt8WamRAt4XKc639wF) 🔗 [资料链接](https:\u002F\u002Fpeople.eecs.berkeley.edu\u002F~pabbeel\u002Fcs287-fa19\u002F)\n\n---\n\n如有任何问题，请通过[Twitter](https:\u002F\u002Ftwitter.com\u002Fomarsar0)联系我。\n\n如果您有意贡献，欢迎提交包含课程链接的PR。虽然这需要一些时间，但我计划对这些单独的讲座进行整理，例如总结内容、添加笔记、提供补充阅读材料、标注难度等。\n\n您现在可以在[这里](https:\u002F\u002Fgithub.com\u002Fdair-ai\u002FML-Course-Notes)找到机器学习课程笔记。","# ML-YouTube-Courses 快速上手指南\n\n**工具简介**：\n`ML-YouTube-Courses` 并非一个需要安装运行的软件库，而是一个由 DAIR.AI 维护的**精选机器学习视频课程索引仓库**。它汇集了全球顶尖高校（如斯坦福、MIT、加州理工）及行业专家（如 Andrej Karpathy、吴恩达）在 YouTube 上发布的最新、最优质的机器学习和深度学习课程。本指南将帮助你快速浏览并访问这些学习资源。\n\n## 环境准备\n\n由于本项目仅为课程链接列表，**无需配置复杂的开发环境或安装依赖**。你只需要：\n\n*   **操作系统**：任意支持现代浏览器的系统（Windows, macOS, Linux）。\n*   **网络环境**：\n    *   能够访问 **YouTube** 的网络环境。\n    *   **国内开发者建议**：由于 YouTube 在中国大陆无法直接访问，建议配置科学上网工具，或寻找部分课程在 **Bilibili (B 站)** 上的官方\u002F非官方搬运镜像（许多经典课程如 CS229, CS231N 在 B 站均有高清中字资源）。\n*   **浏览器**：推荐 Chrome, Edge 或 Firefox。\n\n## 安装步骤（获取资源）\n\n你不需要运行 `pip install` 或 `git clone` 来“使用”这些课程，但你可以克隆仓库以便离线查看目录结构或在本地搜索。\n\n### 方法一：直接在线浏览（推荐）\n直接访问 GitHub 仓库页面查看整理好的课程列表：\n```bash\n# 在浏览器中打开\nhttps:\u002F\u002Fgithub.com\u002Fdair-ai\u002FML-YouTube-Courses\n```\n\n### 方法二：克隆到本地（可选）\n如果你希望保存这份索引清单或在本地检索：\n\n```bash\ngit clone https:\u002F\u002Fgithub.com\u002Fdair-ai\u002FML-YouTube-Courses.git\ncd ML-YouTube-Courses\n```\n\n*注意：克隆后你得到的是包含课程标题和链接的 Markdown 文件，而非视频文件本身。*\n\n## 基本使用\n\n使用流程非常简单：**选择领域 -> 点击链接 -> 开始学习**。\n\n### 1. 选择学习路径\n根据仓库 `README` 中的分类，找到你感兴趣的方向。以下是热门领域及代表课程：\n\n*   **机器学习基础 (Machine Learning)**\n    *   入门首选：`Stanford CS229: Machine Learning` (吴恩达经典)\n    *   理论深入：`Caltech CS156: Learning from Data`\n*   **深度学习 (Deep Learning)**\n    *   代码实战：`Neural Networks: Zero to Hero` (Andrej Karpathy 主讲，从零手写大模型)\n    *   系统课程：`Stanford CS230: Deep Learning`\n*   **大语言模型与应用 (Practical ML \u002F LLMs)**\n    *   应用开发：`LangChain for LLM Application Development`\n    *   提示工程：`ChatGPT Prompt Engineering for Developers`\n    *   全栈实战：`Full Stack LLM Bootcamp`\n*   **自然语言处理 (NLP)**\n    *   核心课程：`CS224N: Natural Language Processing with Deep Learning`\n    *   实战教程：`NLP Course (Hugging Face)`\n*   **计算机视觉 (Computer Vision)**\n    *   经典必修：`CS231N: Convolutional Neural Networks for Visual Recognition`\n\n### 2. 访问课程\n在仓库的 Markdown 文件中，点击对应课程标题后的链接（格式通常为 `🔗 [Link to Course]`），即可跳转至 YouTube 播放列表。\n\n**示例操作**：\n如果你想学习如何从零构建神经网络：\n1.  在文档中定位到 **Deep Learning** 章节。\n2.  找到 `Neural Networks: Zero to Hero by Andrej Karpathy`。\n3.  点击链接进入播放列表。\n4.  按顺序观看 `Lecture 1: The Learning Problem` 等视频。\n\n### 3. 配合代码练习\n大多数课程（如 CS229, CS231N, Zero to Hero）会在其视频描述或课程主页提供配套的代码笔记（Notebooks）和作业。\n*   访问视频下方的描述栏。\n*   查找 \"Course Website\", \"Materials\", 或 \"GitHub\" 链接。\n*   下载代码并在本地 Jupyter Notebook 或 Google Colab 中运行实践。\n\n---\n**提示**：对于国内开发者，若在 YouTube 加载缓慢，可尝试复制视频标题在 **Bilibili** 搜索，通常能找到社区上传的带中文字幕版本，学习效果更佳。","某初创公司的算法工程师李明需要在两周内快速掌握大语言模型（LLM）应用开发技术，以便为公司构建基于私有数据的智能客服原型。\n\n### 没有 ML-YouTube-Courses 时\n- **资源筛选耗时巨大**：在 YouTube 海量视频中盲目搜索，难以区分过时的教程与最新的 LLM 实战课程，浪费大量时间在低质量内容上。\n- **知识体系碎片化**：找到的视频零散分布，缺乏从提示词工程到 LangChain 架构再到向量数据库的系统性学习路径，导致知识结构混乱。\n- **错过顶尖高校资源**：容易忽略斯坦福、MIT 等名校刚发布的深度学习或医疗 AI 前沿课程，只能接触到营销号式的浅层讲解。\n- **验证成本高昂**：无法确认所学技术栈（如 RAG 架构）是否已被工业界验证，可能在错误的方向上投入宝贵的开发时间。\n\n### 使用 ML-YouTube-Courses 后\n- **精准锁定前沿课程**：直接通过\"Practical Machine Learning\"分类找到《LangChain for LLM Application Development》和《Building LLM-Powered Apps》等最新实战教程，立即上手。\n- **构建系统化学习路径**：依托仓库整理的分类索引，按顺序学习从提示词工程到系统构建的完整链条，迅速建立起清晰的工程化思维。\n- **直达顶级学术资源**：一键访问斯坦福 CS25\"Transformers United\"或 MIT 深度学习课程，获取最权威的理论支撑与技术洞察。\n- **提升研发效率**：基于经过社区筛选的高质量内容快速复现方案，将原本需要一个月的调研期压缩至三天，加速原型落地。\n\nML-YouTube-Courses 通过结构化整理全球优质 AI 教育资源，让开发者从“大海捞针”转变为“按需取用”，极大降低了前沿技术的学习门槛与时间成本。","https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Fdair-ai_ML-YouTube-Courses_e704ff6a.png","dair-ai","DAIR.AI","https:\u002F\u002Foss.gittoolsai.com\u002Favatars\u002Fdair-ai_38e7eafe.png","Democratizing Artificial Intelligence Research, Education, and Technologies",null,"dair_ai","https:\u002F\u002Fwww.dair.ai\u002F","https:\u002F\u002Fgithub.com\u002Fdair-ai",17138,2101,"2026-04-03T22:21:52","CC0-1.0",1,"","未说明",{"notes":30,"python":28,"dependencies":31},"该仓库并非可执行的 AI 软件工具，而是一个机器学习课程视频链接的索引列表。用户只需通过浏览器访问提供的 YouTube 链接即可观看课程内容（涵盖机器学习、深度学习、NLP、计算机视觉等），因此无需配置任何本地运行环境、GPU、内存或安装 Python 依赖库。",[],[33,34,35,36,37,38],"开发框架","语言模型","其他","Agent","数据工具","图像",[40,41,42,43,44,45],"machine-learning","deep-learning","nlp","natural-language-processing","ai","data-science",7,"ready","2026-03-27T02:49:30.150509","2026-04-06T07:15:10.606414",[51,56,61,66,71,75],{"id":52,"question_zh":53,"answer_zh":54,"source_url":55},13875,"仓库是否接受较旧的机器学习课程（例如 2015-2016 年发布的）？","目前不接受。维护者决定暂时只收录较新的课程（2020 年、2021 年及以后发布的），以保持仓库内容的时效性。尽管旧课程的核心原理可能仍然适用，但收录标准目前严格限制为新课程。","https:\u002F\u002Fgithub.com\u002Fdair-ai\u002FML-YouTube-Courses\u002Fissues\u002F2",{"id":57,"question_zh":58,"answer_zh":59,"source_url":60},13876,"是否计划添加 YouTube 以外的其他学习平台（如认证课程平台）的链接？","目前没有计划。该项目旨在专注于收集 YouTube 上的公开课程。如果未来有更多用户对此感兴趣，可能会重新考虑，但目前会保持聚焦于 YouTube 资源。","https:\u002F\u002Fgithub.com\u002Fdair-ai\u002FML-YouTube-Courses\u002Fissues\u002F15",{"id":62,"question_zh":63,"answer_zh":64,"source_url":65},13877,"如何获取各章节或讲座的单独笔记？","相关的笔记正在制作中（WIP）。您可以先在独立的仓库中找到部分笔记，地址为：https:\u002F\u002Fgithub.com\u002Fdair-ai\u002FML-Course-Notes","https:\u002F\u002Fgithub.com\u002Fdair-ai\u002FML-YouTube-Courses\u002Fissues\u002F5",{"id":67,"question_zh":68,"answer_zh":69,"source_url":70},13878,"是否推荐 Fast.AI 的“程序员深度学习”课程？","是的，该课程已被收录。这是一门适合从零开始学习应用深度学习的课程，涵盖计算机视觉、NLP、协同过滤等内容，维护者已确认将其添加到列表中。","https:\u002F\u002Fgithub.com\u002Fdair-ai\u002FML-YouTube-Courses\u002Fissues\u002F6",{"id":72,"question_zh":73,"answer_zh":74,"source_url":55},13879,"提交新课程之前需要做什么？","在提交拉取请求（PR）之前，建议先创建一个 Issue 询问维护者是否对该课程感兴趣。如果课程符合收录标准（如发布时间较新），维护者确认后您再提交 PR 会更高效。",{"id":76,"question_zh":77,"answer_zh":78,"source_url":55},13880,"仓库对收录课程的时间范围有什么具体要求？","仓库目前严格限制只收录 2020 年、2021 年及之后发布的课程。这是为了确保学习内容的最新性，即使某些旧课程的 API 变化不大，也不在目前的收录范围内。",[],[81,90,99,107,115,126],{"id":82,"name":83,"github_repo":84,"description_zh":85,"stars":86,"difficulty_score":87,"last_commit_at":88,"category_tags":89,"status":47},3808,"stable-diffusion-webui","AUTOMATIC1111\u002Fstable-diffusion-webui","stable-diffusion-webui 是一个基于 Gradio 构建的网页版操作界面，旨在让用户能够轻松地在本地运行和使用强大的 Stable Diffusion 图像生成模型。它解决了原始模型依赖命令行、操作门槛高且功能分散的痛点，将复杂的 AI 绘图流程整合进一个直观易用的图形化平台。\n\n无论是希望快速上手的普通创作者、需要精细控制画面细节的设计师，还是想要深入探索模型潜力的开发者与研究人员，都能从中获益。其核心亮点在于极高的功能丰富度：不仅支持文生图、图生图、局部重绘（Inpainting）和外绘（Outpainting）等基础模式，还独创了注意力机制调整、提示词矩阵、负向提示词以及“高清修复”等高级功能。此外，它内置了 GFPGAN 和 CodeFormer 等人脸修复工具，支持多种神经网络放大算法，并允许用户通过插件系统无限扩展能力。即使是显存有限的设备，stable-diffusion-webui 也提供了相应的优化选项，让高质量的 AI 艺术创作变得触手可及。",162132,3,"2026-04-05T11:01:52",[33,38,36],{"id":91,"name":92,"github_repo":93,"description_zh":94,"stars":95,"difficulty_score":96,"last_commit_at":97,"category_tags":98,"status":47},1381,"everything-claude-code","affaan-m\u002Feverything-claude-code","everything-claude-code 是一套专为 AI 编程助手（如 Claude Code、Codex、Cursor 等）打造的高性能优化系统。它不仅仅是一组配置文件，而是一个经过长期实战打磨的完整框架，旨在解决 AI 代理在实际开发中面临的效率低下、记忆丢失、安全隐患及缺乏持续学习能力等核心痛点。\n\n通过引入技能模块化、直觉增强、记忆持久化机制以及内置的安全扫描功能，everything-claude-code 能显著提升 AI 在复杂任务中的表现，帮助开发者构建更稳定、更智能的生产级 AI 代理。其独特的“研究优先”开发理念和针对 Token 消耗的优化策略，使得模型响应更快、成本更低，同时有效防御潜在的攻击向量。\n\n这套工具特别适合软件开发者、AI 研究人员以及希望深度定制 AI 工作流的技术团队使用。无论您是在构建大型代码库，还是需要 AI 协助进行安全审计与自动化测试，everything-claude-code 都能提供强大的底层支持。作为一个曾荣获 Anthropic 黑客大奖的开源项目，它融合了多语言支持与丰富的实战钩子（hooks），让 AI 真正成长为懂上",138956,2,"2026-04-05T11:33:21",[33,36,34],{"id":100,"name":101,"github_repo":102,"description_zh":103,"stars":104,"difficulty_score":96,"last_commit_at":105,"category_tags":106,"status":47},2271,"ComfyUI","Comfy-Org\u002FComfyUI","ComfyUI 是一款功能强大且高度模块化的视觉 AI 引擎，专为设计和执行复杂的 Stable Diffusion 图像生成流程而打造。它摒弃了传统的代码编写模式，采用直观的节点式流程图界面，让用户通过连接不同的功能模块即可构建个性化的生成管线。\n\n这一设计巧妙解决了高级 AI 绘图工作流配置复杂、灵活性不足的痛点。用户无需具备编程背景，也能自由组合模型、调整参数并实时预览效果，轻松实现从基础文生图到多步骤高清修复等各类复杂任务。ComfyUI 拥有极佳的兼容性，不仅支持 Windows、macOS 和 Linux 全平台，还广泛适配 NVIDIA、AMD、Intel 及苹果 Silicon 等多种硬件架构，并率先支持 SDXL、Flux、SD3 等前沿模型。\n\n无论是希望深入探索算法潜力的研究人员和开发者，还是追求极致创作自由度的设计师与资深 AI 绘画爱好者，ComfyUI 都能提供强大的支持。其独特的模块化架构允许社区不断扩展新功能，使其成为当前最灵活、生态最丰富的开源扩散模型工具之一，帮助用户将创意高效转化为现实。",107662,"2026-04-03T11:11:01",[33,38,36],{"id":108,"name":109,"github_repo":110,"description_zh":111,"stars":112,"difficulty_score":96,"last_commit_at":113,"category_tags":114,"status":47},3704,"NextChat","ChatGPTNextWeb\u002FNextChat","NextChat 是一款轻量且极速的 AI 助手，旨在为用户提供流畅、跨平台的大模型交互体验。它完美解决了用户在多设备间切换时难以保持对话连续性，以及面对众多 AI 模型不知如何统一管理的痛点。无论是日常办公、学习辅助还是创意激发，NextChat 都能让用户随时随地通过网页、iOS、Android、Windows、MacOS 或 Linux 端无缝接入智能服务。\n\n这款工具非常适合普通用户、学生、职场人士以及需要私有化部署的企业团队使用。对于开发者而言，它也提供了便捷的自托管方案，支持一键部署到 Vercel 或 Zeabur 等平台。\n\nNextChat 的核心亮点在于其广泛的模型兼容性，原生支持 Claude、DeepSeek、GPT-4 及 Gemini Pro 等主流大模型，让用户在一个界面即可自由切换不同 AI 能力。此外，它还率先支持 MCP（Model Context Protocol）协议，增强了上下文处理能力。针对企业用户，NextChat 提供专业版解决方案，具备品牌定制、细粒度权限控制、内部知识库整合及安全审计等功能，满足公司对数据隐私和个性化管理的高标准要求。",87618,"2026-04-05T07:20:52",[33,34],{"id":116,"name":117,"github_repo":118,"description_zh":119,"stars":120,"difficulty_score":96,"last_commit_at":121,"category_tags":122,"status":47},2268,"ML-For-Beginners","microsoft\u002FML-For-Beginners","ML-For-Beginners 是由微软推出的一套系统化机器学习入门课程，旨在帮助零基础用户轻松掌握经典机器学习知识。这套课程将学习路径规划为 12 周，包含 26 节精炼课程和 52 道配套测验，内容涵盖从基础概念到实际应用的完整流程，有效解决了初学者面对庞大知识体系时无从下手、缺乏结构化指导的痛点。\n\n无论是希望转型的开发者、需要补充算法背景的研究人员，还是对人工智能充满好奇的普通爱好者，都能从中受益。课程不仅提供了清晰的理论讲解，还强调动手实践，让用户在循序渐进中建立扎实的技能基础。其独特的亮点在于强大的多语言支持，通过自动化机制提供了包括简体中文在内的 50 多种语言版本，极大地降低了全球不同背景用户的学习门槛。此外，项目采用开源协作模式，社区活跃且内容持续更新，确保学习者能获取前沿且准确的技术资讯。如果你正寻找一条清晰、友好且专业的机器学习入门之路，ML-For-Beginners 将是理想的起点。",84991,"2026-04-05T10:45:23",[38,37,123,124,36,35,34,33,125],"视频","插件","音频",{"id":127,"name":128,"github_repo":129,"description_zh":130,"stars":131,"difficulty_score":87,"last_commit_at":132,"category_tags":133,"status":47},3128,"ragflow","infiniflow\u002Fragflow","RAGFlow 是一款领先的开源检索增强生成（RAG）引擎，旨在为大语言模型构建更精准、可靠的上下文层。它巧妙地将前沿的 RAG 技术与智能体（Agent）能力相结合，不仅支持从各类文档中高效提取知识，还能让模型基于这些知识进行逻辑推理和任务执行。\n\n在大模型应用中，幻觉问题和知识滞后是常见痛点。RAGFlow 通过深度解析复杂文档结构（如表格、图表及混合排版），显著提升了信息检索的准确度，从而有效减少模型“胡编乱造”的现象，确保回答既有据可依又具备时效性。其内置的智能体机制更进一步，使系统不仅能回答问题，还能自主规划步骤解决复杂问题。\n\n这款工具特别适合开发者、企业技术团队以及 AI 研究人员使用。无论是希望快速搭建私有知识库问答系统，还是致力于探索大模型在垂直领域落地的创新者，都能从中受益。RAGFlow 提供了可视化的工作流编排界面和灵活的 API 接口，既降低了非算法背景用户的上手门槛，也满足了专业开发者对系统深度定制的需求。作为基于 Apache 2.0 协议开源的项目，它正成为连接通用大模型与行业专有知识之间的重要桥梁。",77062,"2026-04-04T04:44:48",[36,38,33,34,35]]