P08 Diffusion Policy: Visuomotor Policy Learning via Action Diffusion
Conditional action diffusion, receding-horizon control, and the distinction between sampling and robot time.
Research & Engineering Notes
Ongoing notes on literature, research, and engineering practice.
Reading
Reading directory →Reading notes organized around research questions, methods, experiments, and limitations.
P35 For SALE: State-Action Representation Learning for Deep Reinforcement Learning
SALE objectives, TD7 engineering components, and the scope of empirical evidence.
P36 High-Dimensional Continuous Control Using Generalized Advantage Estimation
TD-residual advantages, reward-to-go, lambda returns, and critic error.
Technical Tutorials
Tutorial directory →18 bilingual tutorials on learning implementation, robot simulation, navigation scenes, VLA, and engineering toolchains.
OpenPI × LIBERO: Serving, Observations, and Action Contracts
Use pinned upstream code to check fields, image orientation, action queues, and a single request.
Robot Assets: CAD, USD, URDF, and Inertial Validation
Validate units, composition layers, joint frames, and physical properties in stages.
PyTorch: Axes, Broadcasting, and Storage Layout
Define shape contracts and distinguish axis changes, reshaping, views, and copies.
Research & Engineering
Browse by category →Theory & Methods
RL foundations, probability, policy optimization, and action modeling.
Benchmarks & Evaluation
Benchmark selection, statistical evaluation, and reproducible protocols.
Research
Research questions, method analysis, and academic discussions.
Engineering
Toolchains, implementation, deployment, and troubleshooting.
Reproduction
Algorithm implementation, experiments, results, and reproducibility.
A Reproducible RL Evaluation and Reporting Protocol
Align environment versions, training budgets, checkpoint selection, held-out testing, and aggregation.
Embodied RL, LLMs/VLAs, and Research Design
Turn sample efficiency, adaptation, rewards, and timing into controlled research questions.
Cross-Paper Comparison and Research Roadmap
Connections and controlled experiments across action generation, policy adaptation, execution, human motion, and navigation worlds.
Hexo and GitHub Pages: Building and Publishing While Preserving Sources
Dependency installation, static generation, browser verification, and source-preserving Git publication.