附录E 参考文献与延伸资源

以下资料主要用于核验本书的技术机制、项目方法和风险边界。网页访问日期为2026年8月6日。具体产品界面、版本和安装步骤不属于纸书稳定内容。

一、人工智能与机器学习基础

  1. Google for Developers. Machine Learning Crash Course[EB/OL]. https://developers.google.com/machine-learning/crash-course.
  2. Google for Developers. Machine Learning Glossary[EB/OL]. https://developers.google.com/machine-learning/glossary.
  3. Google for Developers. Problem Framing: Understand the Problem[EB/OL]. https://developers.google.com/machine-learning/problem-framing/problem.
  4. Google for Developers. Problem Framing: Framing an ML Problem[EB/OL]. https://developers.google.com/machine-learning/problem-framing/ml-framing.
  5. Google for Developers. Implementing a Model[EB/OL]. https://developers.google.com/machine-learning/problem-framing/implement-model.
  6. Google for Developers. Managing ML Projects: Planning, Teams and Pipelines[EB/OL]. https://developers.google.com/machine-learning/managing-ml-projects/.
  7. Google Creative Lab. Teachable Machine[EB/OL]. https://teachablemachine.withgoogle.com/.

二、深度学习、视觉、声音和时序

  1. Keras Team. Keras Code Examples[EB/OL]. https://keras.io/examples/.
  2. Keras Team. Text Classification from Scratch[EB/OL]. https://keras.io/examples/nlp/text_classification_from_scratch/.
  3. TensorFlow. Transfer Learning and Fine-tuning[EB/OL]. https://www.tensorflow.org/tutorials/images/transfer_learning.
  4. TensorFlow. Transfer Learning with YAMNet for Environmental Sound Classification[EB/OL]. https://www.tensorflow.org/tutorials/audio/transfer_learning_audio.
  5. TensorFlow. Time Series Forecasting[EB/OL]. https://www.tensorflow.org/tutorials/structured_data/time_series.
  6. Google AI Edge. LiteRT Models: Build and Convert Models for On-device Runtime[EB/OL]. https://ai.google.dev/edge/litert/models/.
  7. TensorFlow. Data Validation and Responsible AI Resources[EB/OL]. https://www.tensorflow.org/responsible_ai.

三、语言模型、检索与工具连接

  1. Vaswani A, Shazeer N, Parmar N, et al. Attention Is All You Need[C]//Advances in Neural Information Processing Systems. 2017. https://research.google/pubs/attention-is-all-you-need/.
  2. Lewis P, Perez E, Piktus A, et al. Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks[C]//Advances in Neural Information Processing Systems. 2020. https://arxiv.org/abs/2005.11401.
  3. Model Context Protocol. Specification[EB/OL]. https://modelcontextprotocol.io/specification/.

四、数据、模型文档与风险管理

  1. Gebru T, Morgenstern J, Vecchione B, et al. Datasheets for Datasets[J]. Communications of the ACM, 2021, 64(12): 86-92. https://doi.org/10.1145/3458723.
  2. Mitchell M, Wu S, Zaldivar A, et al. Model Cards for Model Reporting[C]//Proceedings of the Conference on Fairness, Accountability, and Transparency. 2019: 220-229. https://research.google/pubs/model-cards-for-model-reporting/.
  3. Tabassi E. Artificial Intelligence Risk Management Framework (AI RMF 1.0)[R]. Gaithersburg: National Institute of Standards and Technology, 2023. https://doi.org/10.6028/NIST.AI.100-1.
  4. Autio C, Schwartz R, Dunietz J, et al. Artificial Intelligence Risk Management Framework: Generative Artificial Intelligence Profile (NIST AI 600-1)[R]. Gaithersburg: National Institute of Standards and Technology, 2024. https://doi.org/10.6028/NIST.AI.600-1.

五、使用建议

机器学习基础、数据划分和评价可从第1—6项开始；视觉、声音、时序和端侧实践可结合第8—14项中的官方Notebook；语言模型、检索和工具连接只需理解第15—17项所描述的基本机制；项目验收、数据卡、模型卡和风险管理可重点参考第18—21项。

教师和学生引用网页代码时，应同时记录页面日期、运行环境和修改内容。示例代码能运行不等于适合当前专业任务，仍须使用本书的任务合同、固定测试、人工接管和交付要求重新验证。

六、完整项目呈现参考

  1. Mistry S. How to Build an ML-powered Doorbell Notifier[EB/OL]. Hackster.io. https://www.hackster.io/sandeep-mistry/how-to-build-an-ml-powered-doorbell-notifier-0a781e.

该案例把项目简介、硬件与软件、数据采集、模型训练、部署运行、测试和完整代码库连成一条可复现链，可用于检查教材章节是否已经成为完整项目。它不是本书项目的固定模板；各章仍须依据自身技术主题决定数据、基线、界面和验收证据。