项目
News: Our slides from the FAccT Tutorial on Responsible AI Toolbox are available here. ML algorithms and systems are often prone to severe bias and highly consequential failure modes that are not well understood. This project advances the methods, tools,…
This projects rethinks the way how VQA models need to trained by proposing to link reasoning questions with simpler, sub questions that are required to learn and solve complex tasks. The work introduces a new dataset, VQA-Introspect with sub questions…
成立:
Developed in a collaboration between Aether, MSR, and Office, these guidelines for human-AI interaction synthesize more than 20 years of thinking and research in human-AI interaction.
The Platform for Situated Intelligence project focuses on constructing an open, extensible platform to enable the fast development of multimodal, integrative-AI systems.
The situated interaction research effort aims to enable computers to reason more deeply about their surroundings, and engage in fluid interaction with humans in physically situated settings. When people interact with each other, they engage in a rich, highly coordinated,…
The importance of intelligibility and transparency in machine learning Most real datasets have hidden biases. Being able to detect the impact of the bias in the data on the model, and then to repair the model, is critical if we…
成立:
Driver attention is a valuable commodity in maintaining driving safety. However, with the proliferation of many interactive devices that place demands on the driver’s attention while driving, effectively allocating attention with the primary goal of managing driving safety presents substantial…
The Windflow project explores the research questions: Could airplanes in flight be employed as a vast sensor network to determine atmospheric conditions? Could data available today be used to infer winds on a large scale without special plane-based wind sensors…
Conversational systems interact with people through language to assist, enable, or entertain. Research at Microsoft spans dialogs that use language exclusively, or in conjunctions with additional modalities like gesture; where language is spoken or in text; and in a variety…