reward-shaping-engineering
Master reward function design - potential-based shaping, hacking patterns, validation
发现 生活与效率 相关 Skills,比较来源仓库和用途,查看适合你工作流的技能说明。
Master reward function design - potential-based shaping, hacking patterns, validation
Use when stakeholders pressure you to change technical priorities and you're tempted to compromise on security-first or call it synthesis - enforces risk-based prioritization over stakeholder preferences
Master ε-greedy, UCB, curiosity-driven, RND, intrinsic motivation exploration
Use when cataloging technical debt under time pressure and tempted to explain choices instead of delivering document - enforces execution discipline with scoped delivery patterns for partial catalogs
Systematic RL debugging - diagnose learning failures, common bugs, diagnosis trees
Master CQL, IQL, BCQ - offline RL from fixed datasets without environment interaction
Master DQN, Double DQN, Dueling DQN, Rainbow - value-based methods for discrete actions
Master REINFORCE, PPO, TRPO - direct policy optimization with trust regions
Master Dyna-Q, MBPO, Dreamer - model-based RL with world models and planning
Validate architecture artifacts against contracts and consistency requirements - codifies systematic validation practices with professional reporting
Generate C4 architecture diagrams from subsystem catalogs - codifies abstraction strategies, notation conventions, and selection criteria
Use when analyzing existing codebases to generate architecture documentation - coordinates subagent-driven exploration with mandatory workspace structure, validation gates, and pressure-resistant workflows
Gym/gymnasium API - custom environments, spaces, wrappers, vectorization, debugging
Master A2C, A3C, SAC, TD3 - actor-critic methods for continuous control
Routes to appropriate deep-RL skills based on problem type and algorithm family
Use when assessing codebase architecture and you feel pressure to soften critique, lead with strengths, or frame problems diplomatically - provides evidence-based critical assessment resisting relationship and...
Master QMIX, MADDPG, CTDE - multi-agent learning with coordination and credit assignment
Analyze unfamiliar codebases systematically to produce subsystem catalog entries - emphasizes strict contract compliance and confidence marking
Master RL theory - MDPs, value functions, Bellman equations, value/policy iteration, TD
Use when you have architecture documentation from system-archaeologist and need critical assessment, refactoring recommendations, or improvement prioritization - routes to appropriate architect specialist skill...
Rigorous RL evaluation - statistical protocols, train/test discipline, metrics, generalization
Synthesize subsystem catalogs and architecture diagrams into comprehensive, navigable architecture reports - codifies synthesis strategies and professional documentation patterns
Create distinctive, production-grade blog posts. Use this skill when the user asks to build a new blog poast
Generate images using Google's Gemini image generation model with Deno. Use this skill when the user wants to create AI-generated images, perform image-to-image transformations, or generate visual content from...
默认按已收录数据中的 GitHub 仓库 Stars 排序。Stars 属于来源仓库,同一仓库的多个技能可能共享相同数值;它反映仓库关注度,不代表单个技能的安装量或质量评分。