Wasserstein 空間深度
本研究提出了一種新的統計深度概念，稱為 Wasserstein 空間深度 (WSD)，用於在 Wasserstein 空間中對分佈數據進行排序和排序。
Wasserstein Spatial Depth: A Novel Measure for Ordering Distributions in Wasserstein Spaces
This paper introduces Wasserstein Spatial Depth (WSD), a novel statistical depth function designed specifically for ordering and ranking distributions within Wasserstein spaces, addressing the limitations of traditional depth measures in this context.
다중 끌개를 가진 시스템을 위한 시간 지연 비선형 맵을 사용한 데이터 기반 모델 식별 및 위상 공간 정보를 활용한 알고리즘 성능 향상
이 연구는 다중 끌개를 가진 시스템의 동역학을 효과적으로 모델링하기 위해 시간 지연 비선형 맵과 위상 공간 정보를 활용한 데이터 기반 시스템 식별 알고리즘(NLDM)을 제시합니다.
Data-Driven Nonlinear System Identification Using Time Delayed Maps with Application to Multi-Attractor Systems
This paper introduces a novel data-driven method called Nonlinear Delayed Maps (NLDM) for identifying nonlinear dynamical systems, particularly those with multiple attractors, by leveraging time-delayed states and nonlinear feature mappings to improve prediction accuracy across different regions of the phase space.
Infomapベースの階層的コミュニティ検出を用いた、都市部のオンデマンド配送における多層的な管理手法
本稿では、Infomapを用いた階層的なコミュニティ検出手法を都市部のオンデマンド配送ネットワークに適用することで、配送効率の向上を実現する多層的な管理フレームワークを提案しています。
A Review of Deep Learning Techniques for Identifying Heavy Particles in Jet Physics, Focusing on Transformer Networks
Deep learning, particularly transformer networks, revolutionizes heavy particle identification in particle physics by effectively analyzing complex jet data, surpassing traditional methods in accuracy and scalability.
Stochastic Quantization and its Connection to Diffusion Models in Machine Learning
This review explores the intriguing connection between stochastic quantization in physics and diffusion models in machine learning, highlighting their shared mathematical framework based on stochastic differential equations (SDEs) and their potential for mutually beneficial applications.
2024 年開放催化劑實驗 (OCx24)：橋接實驗與計算模型 - 以數據驅動的 Sabatier 火山圖預測 HER 催化活性
本文介紹了 2024 年開放催化劑實驗 (OCx24) 數據集，該數據集結合大規模實驗和計算篩選，以構建預測模型，用於加速發現綠色氫氣生產和二氧化碳升級回收的催化劑，特別強調了數據驅動方法在識別有前景的氫析出反應 (HER) 催化劑方面的能力。
2024 개방형 촉매 실험 (OCx24): 실험과 계산 모델의 연결
본 논문에서는 대규모 실험 데이터 세트인 OCx24를 통해 머신러닝 모델을 활용하여 촉매 발견을 가속화하고 계산 모델과 실험 결과 사이의 차이를 해소하는 방법을 제시합니다.
2024年オープン触媒実験（OCx24）：実験と計算モデルの橋渡し - 大規模データセットを用いた触媒性能予測モデルの構築
大規模な実験データセットと計算モデルを組み合わせることで、グリーン水素製造やCO2アップサイクリングに不可欠な、低コストで耐久性があり、効果的な触媒の発見を加速できる。
© 2024 by Linnk AI