63 subscribers
با برنامه Player FM !
Can You Rely on Your AI? Applying the AIR Tool to Improve Classifier Performance
Manage episode 421358557 series 1264075
Modern analytic methods, including artificial intelligence (AI) and machine learning (ML) classifiers, depend on correlations; however, such approaches fail to account for confounding in the data, which prevents accurate modeling of cause and effect and often leads to prediction bias. The Software Engineering Institute (SEI) has developed a new AI Robustness (AIR) tool that allows users to gauge AI and ML classifier performance with unprecedented confidence. This project is sponsored by the Office of the Under Secretary of Defense for Research and Engineering to transition use of our AIR tool to AI users across the Department of Defense. During the webcast, the research team will hold a panel discussion on the AIR tool and discuss opportunities for collaboration. Our team efforts focus strongly on transition and provide guidance, training, and software that put our transition collaborators on a path to successful adoption of this technology to meet their AI/ML evaluation needs.
What Attendees Will Learn:
• How AIR adds analytical capability that didn’t previously exist, enabling an analysis to characterize and measure the overall accuracy of the AI as the underlying environment changes
• Examples of the AIR process and results from causal discovery to causal identification to causal inference • Opportunities for partnership and collaboration
164 قسمت
Manage episode 421358557 series 1264075
Modern analytic methods, including artificial intelligence (AI) and machine learning (ML) classifiers, depend on correlations; however, such approaches fail to account for confounding in the data, which prevents accurate modeling of cause and effect and often leads to prediction bias. The Software Engineering Institute (SEI) has developed a new AI Robustness (AIR) tool that allows users to gauge AI and ML classifier performance with unprecedented confidence. This project is sponsored by the Office of the Under Secretary of Defense for Research and Engineering to transition use of our AIR tool to AI users across the Department of Defense. During the webcast, the research team will hold a panel discussion on the AIR tool and discuss opportunities for collaboration. Our team efforts focus strongly on transition and provide guidance, training, and software that put our transition collaborators on a path to successful adoption of this technology to meet their AI/ML evaluation needs.
What Attendees Will Learn:
• How AIR adds analytical capability that didn’t previously exist, enabling an analysis to characterize and measure the overall accuracy of the AI as the underlying environment changes
• Examples of the AIR process and results from causal discovery to causal identification to causal inference • Opportunities for partnership and collaboration
164 قسمت
همه قسمت ها
×

1 A New Performance Zone for Software for National Security 1:02:23









1 Independent Verification and Validation for Agile Projects 1:02:23

1 Generative AI and Software Engineering Education 1:02:05
به Player FM خوش آمدید!
Player FM در سراسر وب را برای یافتن پادکست های با کیفیت اسکن می کند تا همین الان لذت ببرید. این بهترین برنامه ی پادکست است که در اندروید، آیفون و وب کار می کند. ثبت نام کنید تا اشتراک های شما در بین دستگاه های مختلف همگام سازی شود.