Microsoft Research Blog

Research Blog

  1. graphical user interface, text, application, email

    The science behind semantic search: How AI from Bing is powering Azure Cognitive Search 

    March 2, 2021

    Azure Cognitive Search (opens in new tab) is a cloud search service that gives developers APIs and tools to build rich search experiences over private, heterogeneous content in web, mobile, and enterprise applications. It has multiple components, including an API for indexing and querying, seamless integration through Azure data ingestion, deep integration with Azure Cognitive Services, and persistent storage…

  2. hexagonal AI icons on a blue background

    AAAI 2021: Accelerating the impact of artificial intelligence 

    February 24, 2021

    The purpose of the Association for the Advancement of Artificial Intelligence, according to its bylaws, is twofold. The first is to promote research in the area of AI, and the second is to promote the responsible use of these types of technology. The result was a 35th AAAI Conference on Artificial Intelligence (AAAI-21) (opens in new tab) schedule that broadens the…

  3. Designer-centered reinforcement learning 

    February 17, 2021 | Batu Aytemiz, Mikhail Jacob, Sam Devlin, and Katja Hofmann

    In video games, nonplayer characters, bots, and other game agents help bring a digital world and its story to life. They can help make the mission of saving humanity feel urgent, transform every turn of a corner into a gamer’s potential demise, and intensify the…

  4. LAMBDA graphic

    LAMBDA: The ultimate Excel worksheet function 

    January 25, 2021 | Andy Gordon and Simon Peyton Jones

    Ever since it was released in the 1980s, Microsoft Excel has changed how people organize, analyze, and visualize their data, providing a basis for decision-making for the millions of people who use it each day. It’s also the world’s most widely used programming language. Excel…

  5. diagram

    Three mysteries in deep learning: Ensemble, knowledge distillation, and self-distillation 

    January 19, 2021 | Zeyuan Allen-Zhu and Yuanzhi Li

    Under now-standard techniques, such as over-parameterization, batch-normalization, and adding residual links, “modern age” neural network training—at least for image classification tasks and many others—is usually quite stable. Using standard neural network architectures and training algorithms (typically SGD with momentum), the learned models perform consistently well,…