TrustBuilder.AI: fast, robust, and explainable deep learning

TrustBuilder.AI：快速、稳健且可解释的深度学习

• 批准号: 568643-2021
• 负责人: Hemmati, HadiH
• 金额: $ 3.64万
• 依托单位: University of Calgary
• 依托单位国家: 加拿大
• 项目类别: Alliance Grants
• 财政年份: 2022
• 资助国家: 加拿大
• 起止时间: 2022-01-01 至 2023-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=758033
• 关键词: TrustBuilder.AI; fast; robust; explainable; deep

With a wealth of unlimited AI startups, Canada is now ranked fifth in the world for commercial AI-focused ventures. However, according to a survey by MIT Management Review, 70% of companies surveyed, report minimal or no impact from AI. The main gaps in the market are: (a) the majority of the AI prototypes developed in the market are not production-ready (not scalable to real-world demands) and (b) they have not been able to gain their users' trust on the AI product yet. This proposal's objective is to provide an integrated software solution (TrustBuilder.ai) that helps AI businesses overcome these challenges using a combination of novel techniques from RobustAI, MLOps, and ExplainableAI domains. The proposed solution will be useful for a wide range of enterprises from large-scale ML service providers to SME's in IoT, fintech, and healthcare, all around the world. The tool will be useful for both ML developers, data scientists, and QA engineers in the company as well executive managers and domain experts.

凭借大量无限的人工智能初创公司，加拿大现在在以商业人工智能为重点的企业中排名第五。然而，根据麻省理工学院管理评论的一项调查，70%的受访公司表示人工智能的影响很小或没有影响。市场上的主要差距是：（a）市场上开发的大多数人工智能原型都没有生产就绪（无法扩展到现实世界的需求），（B）他们还无法获得用户对人工智能产品的信任。该提案的目标是提供一个集成的软件解决方案（TrustBuilder.ai），帮助人工智能企业使用RobustAI、MLOps和ExplainableAI领域的新技术组合来克服这些挑战。拟议的解决方案将适用于从大型ML服务提供商到全球物联网、金融科技和医疗保健领域的中小企业的各种企业。该工具对公司的ML开发人员、数据科学家和QA工程师以及执行经理和领域专家都很有用。