Tag: #deep-learning

한국어English日本語中文

← All posts

• The Complete Autonomous Driving & Robotics Tech Stack: From C++, ROS2, CUDA, TensorRT to VLM/VLA, Simulation, and Beyond

  A comprehensive guide to the core technology stack behind autonomous driving and robotics. Covering Modern C++, ROS/ROS2, CUDA parallel programming, TensorRT optimization, model compression (quantization/pruning), sensor

  2026-03-01 · 22 min read #autonomous-driving#robotics#ros2#cuda#tensorrt

• CUDA Hands-on Complete Guide: Everything About GPU Computing

  A comprehensive guide covering NVIDIA CUDA from fundamental concepts to GPU architecture, C/C++ kernel programming, memory optimization, Python integration (PyTorch, Numba, CuPy), Multi-GPU training, profiling, and troub

  2026-03-01 · 51 min read #cuda#gpu#nvidia#deep-learning#machine-learning

• The New Throne Beyond vLLM: 5 Reasons SGLang Is Changing the LLM Inference Landscape

  From SGLang's RadixAttention-based KV cache management and 29% throughput advantage through hyper-specialized design, to its 4,000-line Python zero-overhead scheduler, Prefill-Decode Disaggregation, and Compressed FSM-ba

  2026-03-01 · 35 min read #sglang#llm-inference#vllm#radix-attention#model-serving

• HunyuanVideo and LTX-2 Complete Analysis: Architecture, Performance, and Practical Guide to Open-Source Video Generation Models

  A deep analysis of Tencent HunyuanVideo (13B) and Lightricks LTX-2 (19B) architectures, training methodologies, and performance benchmarks, along with a comprehensive comparison of the open-source video generation ecosys

  2026-03-01 · 30 min read #hunyuan-video#ltx-video#ltx2#text-to-video#image-to-video

• ACE-Step: A New Paradigm in AI Music Generation — Complete Analysis of Architecture, Training Methods, and Practical Applications

  An in-depth analysis of the ACE-Step music generation model covering its architecture, training methodology, and text-to-music generation principles, with comparisons against competing models like MusicGen, Suno, and Udi

  2026-03-01 · 38 min read #ace-step#music-generation#ai-music#deep-learning#diffusion

• Reinforcement Learning Complete Guide: From Theory to the Latest Algorithms and Real Implementations

  Derives the major algorithms with their mathematics — from the core theory of MDPs and the Bellman equations through Q-Learning, DQN, Policy Gradient, A3C, PPO and SAC — and pulls together RLHF, AlphaGo and robotics appl

  2026-03-01 · 46 min read #reinforcement-learning#deep-learning#dqn#ppo#rlhf

• Complete Analysis of the GPT Series Papers: The Journey from GPT-1 to GPT-4, How Language Models Changed the World

  A complete generation-by-generation analysis of OpenAI GPT series. From GPT-1 unsupervised pre-training, GPT-2 zero-shot learning, GPT-3 in-context learning and scaling laws, InstructGPT RLHF, to GPT-4 multimodal capabil

  2026-03-01 · 36 min read #gpt#openai#language-model#transformer#pre-training

• GAN Paper Deep Dive: How Generative Adversarial Networks Ushered in the Era of AI-Generated Content

  From Ian Goodfellow's original GAN paper through DCGAN, WGAN, Progressive GAN, and StyleGAN -- we analyze the core theory of generative adversarial networks including the minimax game, Nash equilibrium, and training inst

  2026-03-01 · 34 min read #gan#generative-model#adversarial-training#deep-learning#image-generation

• Attention Is All You Need - A Complete Analysis of the Transformer Paper

  A thorough analysis of the "Attention Is All You Need" paper, the foundation of the Transformer architecture, breaking down its core mechanisms including Self-Attention, Multi-Head Attention, and Positional Encoding one

  2026-03-01 · 18 min read #ai-papers#transformer#deep-learning#attention

• The Complete Guide to vLLM & Ollama: LLM Serving Engine Setup, Parameters, and Environment Variables

  A comprehensive deep dive comparing vLLM PagedAttention architecture and Ollama local LLM runtime environments. Covers everything from installation, server startup, API calls, key CLI arguments, sampling parameters, envi

  2026-03-01 · 46 min read #vllm#ollama#model-serving#inference#gpu

• OpenAI Codex Complete Analysis: From the Birth of AI Code Generation to the Evolution of Cloud Coding Agents

  A complete analysis of OpenAI Codex covering the history of AI code generation and the evolution toward cloud-based coding agents.

  2026-03-01 · 32 min read #codex#openai#code-generation#github-copilot#ai-coding

• Complete Guide to GPU Memory Optimization and Mixed Precision Training

  Analyze GPU memory components based on NVIDIA official documentation and cover memory optimization techniques including Mixed Precision Training and Gradient Checkpointing.

  2026-03-01 · 22 min read #gpu#cuda#mixed-precision#optimization#deep-learning

• Mastering the Segment Anything Model: Paper Analysis and Practical Guide from SAM 1 to SAM 2 to SAM 3

  A comprehensive deep dive into Meta AI's Segment Anything Model (SAM) series. Covering SAM 1 (image promptable segmentation), SAM 2 (real-time video segmentation), and SAM 3 (concept-aware segmentation) — including archi

  2026-03-01 · 21 min read #sam#segment-anything#computer-vision#image-segmentation#video-segmentation

• ResNet Paper In-Depth Analysis: How Residual Connections Broke the Depth Barrier in Deep Learning

  An in-depth analysis of Microsoft Research ResNet paper. We cover the discovery of the degradation problem, the mathematical principles behind skip connections, bottleneck architecture, and the ImageNet ILSVRC 2015 winni

  2026-03-01 · 29 min read #resnet#residual-learning#cnn#computer-vision#image-classification

• Complete Analysis of the DDPM Paper: The Mathematics and Principles of Diffusion Models that Create Images from Noise

  An in-depth analysis of the DDPM paper by Ho et al. We derive the Forward/Reverse diffusion process, variational lower bound (ELBO), noise scheduling, and simplified training objective from formulas, and provide a compre

  2026-03-01 · 31 min read #ddpm#diffusion-model#generative-model#score-matching#stable-diffusion

• Complete Analysis of the BERT Paper: How Bidirectional Transformers Changed the Landscape of NLP

  An in-depth analysis of Google BERT paper. We cover bidirectional pre-training through Masked Language Model (MLM) and Next Sentence Prediction (NSP), fine-tuning strategies, and the core principles of the architecture t

  2026-03-01 · 28 min read #bert#nlp#transformer#pre-training#fine-tuning

• Vision Transformer (ViT) Paper In-Depth Analysis: An Image is Worth 16x16 Words

  An in-depth analysis of Google ViT paper. We cover the innovative approach of converting images into patch sequences, the principles behind Patch Embedding and Position Embedding, the performance vs. data efficiency trad

  2026-03-01 · 30 min read #vit#vision-transformer#computer-vision#transformer#image-classification