Skip to content

References: The AI HAT+ and Neural Network Fundamentals

  1. Convolutional neural network - Wikipedia - Explains how sliding filters, feature maps, and pooling layers let CNNs detect visual patterns regardless of position, directly grounding this chapter's convolution and image-classification concepts.

  2. Object detection - Wikipedia - Surveys how computer vision systems locate and classify multiple objects within an image, covering the bounding-box and class-label terminology this chapter uses to describe detection output.

  3. Overfitting - Wikipedia - Describes how a model can fit training data too closely and lose the ability to generalize, the exact failure mode this chapter warns against when comparing training and validation accuracy.

  4. Neural Networks and Deep Learning - Michael A. Nielsen - Determination Press (2015) - Nielsen is widely credited for deriving backpropagation through four clearly numbered "fundamental equations," an unusually transparent walk from a single neuron's weighted sum to how a whole network learns, echoing this chapter's own worked example.

  5. Deep Learning - Ian Goodfellow, Yoshua Bengio, and Aaron Courville - MIT Press (2016) - No single author is definitively credited with originating the convolutional-network teaching analogies used industry-wide, so this entry instead names the textbook most consistently recommended in university courses for its unusually clear chapter on sparse connectivity, parameter sharing, and pooling.

  6. Introduction to Deep Learning (6.S191) - MIT OpenCourseWare - MIT's introductory deep learning course materials covering how neural networks are built and trained, giving foundational context for this chapter's weighted-sum, neuron, and layer vocabulary.

  7. Introduction to Convolution Neural Network - GeeksforGeeks - A worked tutorial on how convolutional filters slide across an image to build feature maps, reinforcing this chapter's explanation of why CNNs suit image classification and object detection.

  8. Evaluating Object Detection Models: Methods and Metrics - GeeksforGeeks - Explains bounding boxes, confidence scores, precision, recall, and false positives/negatives with runnable code, directly matching this chapter's vocabulary for judging a detection model's output.

  9. Underfitting and Overfitting in Machine Learning - GeeksforGeeks - Contrasts a model that memorizes training data with one that generalizes well, illustrating the training-versus-validation accuracy gap this chapter uses to define overfitting.

  10. AI HAT+ Documentation - Raspberry Pi Foundation - Official documentation for the AI HAT+ accelerator, detailing the Hailo-8L (13 TOPS) and Hailo-8 (26 TOPS) neural processing units this chapter introduces as dedicated inference hardware.