The National Science Foundation (NSF) awarded a $599,410 Project Grant to Carnegie Mellon University (CMU) to develop a collaborative learning framework for dynamic and diverse computing environments, particularly focused on edge devices. The grant, under NSF's Computer and Information Science and Engineering program (CFDA 47.070), aims to innovate model-parallel collaborative learning by designing unique model architectures and efficient algorithms, facilitate practical on-device training and communication algorithms, and evaluate the trustworthiness and mechanism design of the system in a decentralized setting. The project seeks to improve collaborative learning by addressing challenges like free-riding and mismatches between model architecture and device capabilities, with applications in areas such as disaster forecasting, AI-assisted clinical diagnosis and treatment, and decentralized strategic decision-making. The award is effective from September 1, 2023, to August 31, 2026, and does not include any planned sub-awards.