Mohlo by se vám také líbit

Informovat o naskladnění knihy

Více informací o knize Artificial Intelligence

• Parametry knihy
• Anotace
• Oblíbené z jiného soudku

Nákupem získáte 402 bodů

Anotace knihy

Master Modern Artificial Intelligence from Fundamental Mathematics to Production LLMs.

Most books on Artificial Intelligence fall into two extremes: superficial overviews that treat models as black boxes, or dense academic treatises with zero production code. Artificial Intelligence: From Zero to Mastery by Karl Gill dismantles this divide. It delivers a comprehensive, first-principles journey from fundamental mathematics through classical algorithms and deep neural networks to state-of-the-art Large Language Models, autonomous agent swarms, and production serving.

Whether you are a software engineer transitioning into AI, a data scientist seeking deep architectural mastery, or a technical leader designing enterprise systems, this 135-page master handbook provides the exact mathematical derivations, PyTorch code, and systems engineering blueprints required to excel.

What You Will Master Inside:

• The Four Mathematical Pillars: Linear algebra (SVD, tensors), multivariate calculus (Jacobians, Hessians), probability (Bayes' rule, MLE), and information theory (entropy, KL divergence).
• Classical Machine Learning: Bias-variance tradeoff, L1/L2 regularization geometry, Decision Trees, Random Forests, XGBoost, and Support Vector Machines with the Kernel Trick.
• The Deep Learning Engine: Rigorous backpropagation on computational graphs, activation dynamics (GELU, Swish), advanced optimizers (AdamW), and normalization (LayerNorm, RMSNorm).
• Computer Vision & Spatial Intelligence: Convolutions, ResNet gradient highways, object detectors (YOLO, Faster R-CNN, Focal Loss), U-Net segmentation, and Vision Transformers (ViT).
• NLP & Sequence Modeling: BPE tokenization, Word2Vec, LSTM constant error carousels, GRU gating, and the Bahdanau Attention breakthrough.
• The Transformer Architecture: In-depth mathematical anatomy of scaled dot-product attention, multi-head subspaces, Rotary Embeddings (RoPE), and SwiGLU networks.
• Large Language Models & Scaling: Autoregressive pre-training, Chinchilla scaling laws, Grouped-Query Attention (GQA), FlashAttention-1/2/3, and distributed 3D parallelism (ZeRO-3, FSDP).
• Post-Training Alignment: SFT with loss masking, LoRA/QLoRA 4-bit NormalFloat mechanics, RLHF with PPO, Direct Preference Optimization (DPO), and test-time compute (CoT, PRMs).
• Enterprise RAG Systems: Non-parametric memory, chunking strategies, dense embeddings (InfoNCE), HNSW vector graph indexing, HyDE, hybrid BM25 search, Cross-Encoder re-ranking, and the RAG Triad.
• Autonomous AI Agents: Perception-action loops, ReAct frameworks, Reflexion self-correction, grammar-constrained GBNF decoding, and multi-agent state graphs.
• Production AI Engineering: Prefill vs. decode bottlenecks, INT8/INT4/FP8 quantization (AWQ, GPTQ, GGUF), vLLM PagedAttention, speculative decoding, and prompt injection defense.
• Frontiers of AI & AGI: Multimodal vision-language fusion (CLIP, LLaVA), Diffusion Transformers (DiT), JEPA world models, embodied robotics (VLA), ARC-AGI, and superalignment.

Every chapter concludes with Key Architectural Takeaways and Practice Exercises. Includes standalone appendices with the Mathematical Foundations Cheat Sheet, Production AI Toolchain, and a Comprehensive 100-Term Glossary.

Scroll up, click "Buy Now", and master modern artificial intelligence today!

Parametry knihy

• Plný název: Artificial Intelligence
• Podnázev: From Zero to Mastery: he Definitive Guide to Foundations, Neural Networks, Transformers, Large Language Models, Agents, and Production Systems
• Autor: Karl Gill
• Jazyk: Angličtina
• Vazba: Brožovaná
• Počet stran: 138
• EAN: 9798178813355
• ID: 54075302
• Nakladatelství: Independently published
• Hmotnost: 196 g
• Rozměry: 229 × 152 × 8 mm
• Datum vydání: 04. October 2026

---

---

---