The proliferation of multi-sided platforms (MSPs) in business-to-business (B2B) transactions has fundamentally transformed commercial ecosystems, creating complex networks where value creation emerges from strategic interactions between multiple participant groups. This research presents a comprehensive simulation model designed to analyze and predict network effects within B2B multi-sided platforms, addressing critical gaps in understanding how platform dynamics influence transaction efficiency, participant retention, and ecosystem growth. The study develops a sophisticated mathematical framework that integrates game-theoretic principles, network topology analysis, and agent-based modeling to simulate the emergence and sustainability of network effects in B2B environments. Our research methodology employs a mixed-methods approach combining quantitative simulation techniques with qualitative case study analysis across diverse B2B platform ecosystems. The simulation model incorporates key variables including participant heterogeneity, transaction costs, switching barriers, platform governance mechanisms, and competitive dynamics to generate realistic scenarios of network evolution. Through extensive computational experiments and validation against real-world platform data, we demonstrate the model's capacity to predict critical tipping points, identify optimal platform strategies, and forecast long-term ecosystem sustainability. The findings reveal that network effects in B2B multi-sided platforms exhibit distinct characteristics compared to consumer-oriented platforms, with stronger emphasis on trust-building mechanisms, specialized service integration, and institutional relationship dynamics. The simulation model successfully identifies optimal participant acquisition strategies, reveals the criticality of early-stage platform governance decisions, and quantifies the impact of competitive entry on established platform ecosystems. Our results indicate that successful B2B platforms require sophisticated balancing mechanisms to manage power asymmetries between different participant groups while maintaining sufficient incentives for continued platform engagement. The research contributes to both theoretical understanding and practical application by providing platform operators, investors, and policymakers with data-driven insights for strategic decision-making. The simulation framework offers predictive capabilities for scenario planning, risk assessment, and platform design optimization. Furthermore, the study establishes a foundation for future research in platform economics, network theory, and digital business model innovation within B2B contexts.
ТЕОРІЯ ТИПІВ ГОМОТОПІЙ (HoTT): ЗВ'ЯЗОК МІЖ МАТЕМАТИКОЮ ТА КОМП'ЮТЕРНИМИ НАУКАМИАнотація.Теорія типів гомотопій (ТТГ, Homotopy Type Theory (HoTT))це новаторський підхід, який об'єднує поняття з математики та інформатики (інформаційних технологій), пропонуючи нове розуміння обох галузей.В даній статті поглиблено досліджено ТТГ, висвітлюючи її фундаментальні принципи, історичний розвиток та наслідки для математичних міркувань і обчислювальних методів.Проведено заглиблення в основні поняття теорії гомотопії, теорії типів та їх інтеграції, демонструючи симбіотичний зв'язок між математикою та комп'ютерними науками через призму HoTT.Крім того, обговорюється застосування HoTT в різних областях, від формальної верифікації до топологічного аналізу даних, підкреслюючи його трансформаційний потенціал у
Molecular representation learning plays a crucial role in AI-assisted drug discovery research. Encoding 3D molecular structures through Euclidean neural networks has become the prevailing method in the geometric deep learning community. However, the equivariance constraints and message passing in Euclidean space may limit the network expressive power. In this work, we propose a Harmonic Molecular Representation learning (HMR) framework, which represents a molecule using the Laplace-Beltrami eigenfunctions of its molecular surface. HMR offers a multi-resolution representation of molecular geometric and chemical features on 2D Riemannian manifold. We also introduce a harmonic message passing method to realize efficient spectral message passing over the surface manifold for better molecular encoding. Our proposed method shows comparable predictive power to current models in small molecule property prediction, and outperforms the state-of-the-art deep learning models for ligand-binding protein pocket classification and the rigid protein docking challenge, demonstrating its versatility in molecular representation learning.
Federated Learning coordinates multiple clients to collaboratively train a shared model while preserving data privacy. However, the training data with noisy labels located on the participating clients severely harm the model performance. In this paper, we propose FedCoop, a cooperative Federated Learning framework for noisy labels. FedCoop mainly contains three components and conducts robust training in two phases, data selection and model training. In the data selection phase, in order to mitigate the confirmation bias caused by a single client, the Loss Transformer intelligently estimates the probability of each sample’s label to be clean through cooperating with the helper clients, which have high data trustability and similarity. After that, the Feature Comparator evaluates the label quality for each sample in terms of latent feature space in order to further improve the robustness of noisy label detection. In the model training phase, the Feature Matcher trains the model on both the noisy and clean data in a semi-supervised manner to fully utilize the training data and exploits the feature of global class to increase the consistency of pseudo labeling across the clients. The experimental results show FedCoop outperforms the baselines on various datasets with different noise settings. It effectively improves the model accuracy up to 62% and 27% on average compared with the baselines.
