The Staged Pandora's Box: Reservation Indices for Multi-Fidelity Bayesian Optimization
A staged search must decide which candidate to advance next and when to stop. Some observations only inform la
- 用途
- 技術検証・論文読解補助
- 難易度
- Hard
- コスト
- Medium
Manufacturing AI / Materials Informatics
論文やGitHubを探すだけでなく、配合最適化、品質予測、異常検知、材料探索を自社のExcel/CSVデータで試す最初の一歩まで整理します。
製造業や材料開発では実験数が限られます。LightGBM、Gaussian Process、ベイズ最適化など、nが小さい段階でも試しやすい手法を優先します。
1行1実験、1行1ロットの表形式データを出発点にして、品質予測、配合最適化、異常検知へつなげます。
精度だけでなく、SHAPや特徴量重要度で現場・上司・研究チームに説明できることを重視します。
樹脂、電池、触媒、合金などの配合・組成を、少ない実験回数で改善するためのAI/MI導線です。
LightGBM + SHAPから始める →
工程条件、原料ロット、測定値から、品質指標や不良リスクを予測するための実装ガイドです。
LightGBM分類/回帰から始める →
正常データが中心の工程・設備データから、外れ値や異常兆候を見つけるための導線です。
Isolation Forestから始める →
材料候補、分子、結晶、組成から、次に評価すべき候補を絞り込むためのMIガイドです。
記述子 + LightGBMから始める →
製品画像、顕微鏡画像、表面欠陥を対象に、分類・検出・セグメンテーションを選ぶための導線です。
良品学習型異常検知または画像分類から始める →
過去の実験ノート、技術報告書、特許、論文を探しやすくするための製造業向けRAG導線です。
BM25 + embedding検索から始める →
A staged search must decide which candidate to advance next and when to stop. Some observations only inform la
Treating morphology and control as separate design problems has constrained the efficiency and adaptability of
Detecting defects in real-world industrial inspection often requires learning from only a few normal images pe
Power-grid customer service increasingly relies on community social media, where customers describe outages, b
Traditional image cropping assumes the photograph already contains the composition worth keeping, and asks onl
We present GazeTags, a camera- and computer-vision-free approach for worn head-gaze object awareness. Our smal
Continual adaptation requires deciding not only how to update a model, but whether to commit an update without
Video games provide a practical setting for collecting vision-language-action (VLA) data, whose acquisition in
Deep learning models for electronic health records achieve strong predictive performance but provide little in
As general-purpose Large Language Models (LLMs), such as Gemini and DeepSeek, exhibit increasingly advanced re
LaBraM, an early EEG foundation model, demonstrated that large-scale EEG pretraining can yield strong represen
Computational drug discovery offers a growing range of tools for proposing and evaluating molecules, but combi