AI for Programming Education
Education needs technology-based disruption for personalization. Programming education needs most attention, given the increasing skills gap in the IT industry and the need to introduce computational thinking in schools. Programming education is also challenging as…
Research talk: Approximate nearest neighbor search systems at scale
Building deep learning-based search and recommendation systems at internet scale requires a complete redesign of the search index. Key to this redesign is a fast, accurate, and cost-efficient indexing system for approximate nearest neighbor search.…
Panel: Developer velocity and productivity
The socio-technical nature of the software development process has continually taken a prominent role in how software systems has evolved. Microsoft, as the global leader in software development, is home to the world’s largest developer…
Lightning talks: Augmented Mathematical Intelligence
Augmented Mathematical Intelligence (AMI) refers to the ability to solve formally specified problems as reliably as humans can over all classes of such problems that are of interest to our civilization. This includes not only…
Opening remarks: Empowering software developers and mathematicians with next-generation AI
Machine learning systems have become increasingly capable of making fast, plausible predictions in diverse situations. Combining this capability with the reliability of symbolic reasoning will produce next generation AI that will empower developers and mathematicians…
Panel: Preparing for the next generation of developer communities
As more people are programming for a purpose, they have turned to online communities and resources to grow their developer skills. But there are a few open questions about how people are engaging in these…
Closing remarks: Empowering software developers and mathematicians with next-generation AI
Machine learning systems have become increasingly capable of making fast, plausible predictions in diverse situations. Combining this capability with the reliability of symbolic reasoning will produce next generation AI that will empower developers and mathematicians…
Research talks: AI for software development
This session will showcase a future in which AI automates much of the tedious and error-prone collaborative software development tasks. We take a broad, end-to-end view of software development that consists of not only coding…
Research talk: Torchy: A tracing JIT compiler for PyTorch
This session will introduce current technologies to speed up training of machine learning models and their limitations, and it will also include a presentation of Torchy, a transparent solution to speed up PyTorch workloads through…
Research talk: SPTAG++: Fast hundreds of billions-scale vector search with millisecond response time
Current state-of-the-art vector approximate nearest neighbor search (ANNS) libraries mainly focus on how to do fast high-recall search in memory. However, extremely large-scale vector search scenarios present certain challenges. For example, hundreds of billions of…