Skip to content

搜索技能.../

Agent Skill Search Engine

搜索

搜索
分类
职业

关于

关于
隐私政策
服务条款

© 2026 Skills Pool. 保留所有权利。

Quality Diversity Filtering | Skills Pool

技能档案

Quality Diversity Filtering

MAP-Elites style filter that retains the best idea per gap×method-type niche to maximize coverage

Pthahnix50 星标2026年4月10日

职业
分类: 机器学习

技能内容

Quality-Diversity Filtering SOP

Layer Rules

Agent: dare-agents MCP tool quality_diversity_filter
Called by: review tactic (P1), idea-generation tactic (P2)
Never calls: other SOPs (leaf node)

Procedure

Receive ideas (Idea[] JSON) and gaps (Gap[] JSON) from calling tactic
Call quality_diversity_filter({ ideas, gaps }) via dare-agents
Parse response into QdFilterResult
Return filtered set to calling tactic; removed ideas are not passed forward

Key Behavior

Niche definition: each niche = one gap × one method-type (e.g., gap-3 × contrastive-learning)
Selection rule: within each niche, keep only the highest-quality idea; discard the rest

相关技能

快速安装

Quality Diversity Filtering

npx skillvault add Pthahnix/pthahnix-de-anthropocentric-research-engine-skills-sop-quality-diversity-filtering-skill-md

下载 Skill 打开源码仓库

作者: Pthahnix
星标: 50
更新时间: 2026年4月10日
职业

本页内容

01Quality-Diversity Filtering SOP

Goal: prevent a single approach from dominating; ensure coverage across both gaps and methods

Mirrors MAP-Elites quality-diversity algorithm: quality = score, diversity = niche occupancy

Output Schema

filteredIdeas[] — surviving ideas after niche-best selection, one per occupied niche
removedCount — number of ideas eliminated as niche duplicates
nicheMap — mapping of niche key (gapId + methodType) to the winning idea ID in that niche
reasoning — explanation of key selection decisions, especially contested niches

02

Layer Rules

05Output Schema

Mathematical Science Teachers, Postsecondary

Continuous Learning V2

Instinct-based learning system that observes sessions via hooks, creates atomic instincts with confidence scoring, and evolves them into skills/commands/agents. v2.1 adds project-scoped instincts to prevent cross-project contamination.

Continuous Learning V2

Hook'lar aracılığıyla oturumları gözlemleyen, güven skorlaması ile atomik instinct'ler oluşturan ve bunları skill/command/agent'lara evriltiren instinct tabanlı öğrenme sistemi. v2.1 çapraz proje kontaminasyonunu önlemek için proje kapsamlı instinct'ler ekler.

Continuous Learning V2

훅을 통해 세션을 관찰하고, 신뢰도 점수가 있는 원자적 본능을 생성하며, 이를 스킬/명령어/에이전트로 진화시키는 본능 기반 학습 시스템. v2.1에서는 프로젝트 간 오염을 방지하기 위한 프로젝트 범위 본능이 추가되었습니다.

Continuous Learning

Claude Codeセッションから再利用可能なパターンを自動的に抽出し、将来の使用のために学習済みスキルとして保存します。

Continuous Learning

Automatically extract reusable patterns from Claude Code sessions and save them as learned skills for future use.

Pytorch Patterns

PyTorch deep learning patterns and best practices for building robust, efficient, and reproducible training pipelines, model architectures, and data loading.

Mathematical Science Teachers, Postsecondary