Artificial Intelligence
Building intelligent machines that complement human reasoning and extend what people can do.
Vidéo
Research talk: Privacy in machine learning research at Microsoft
Speaker: Melissa Chase, Principal Researcher, Microsoft Research Redmond Training modern machine learning models requires large amounts of data, and often that data may be private or confidential. The area of privacy-preserving machine learning looks at…
Vidéo
Panel: Generalization in reinforcement learning
The ability for a reinforcement learning (RL) policy to generalize is a key requirement for the broad application of RL algorithms. This generalization ability is also essential to the future of RL—both in theory and…
Vidéo
Demo: RAI Toolbox: An open-source framework for building responsible AI
Assessing and investigating machine learning (ML) models prior to deployment remains at the core of developing trustworthy and responsible artificial intelligence (AI). While different open-source tools have been proposed for assessing fairness, explainability, or errors…
Vidéo
Research talk: Project Dexter: Machine learning and automatic decision-making for robotic manipulation
Robot technology has long held the promise of disrupting many important industries that involve dexterous object manipulation in weakly structured environments, including healthcare, agriculture, and infrastructure maintenance. The increasing versatility of robotic manipulation hardware seemingly…
Vidéo
Tutorial, Research talk, and Q&A: ElectionGuard: Enabling voters to verify election integrity
00:00 Tutorial: ElectionGuard: Enabling voters to verify election integrity Speaker: Josh Benaloh, Senior Principal Cryptographer, Microsoft Research Redmond This tutorial will describe Microsoft’s free, open-source ElectionGuard toolkit, which can be incorporated into new and existing…
Vidéo
Demo: EverParse: Automatic generation of formally verified secure parsers for cloud integrity
Speaker: Tahina Ramananandro, Principal Research Software Development Engineer, Microsoft Research Redmond DARPA and MITRE estimate that 80 percent of software security vulnerabilities have incorrect input validation as their root cause. In such scenarios, attackers provide…
Vidéo
Research talk: NUWA: Neural visual world creation with multimodal pretraining
Over the past years, large-scale pretrained models with billions of parameters have improved the state of the art in nearly every natural language processing (NLP) task. These models are fundamentally changing the research and development…
Vidéo
Keynote: Universal search and recommendation
Search and recommendation are at the heart of how information workers deal with information overload. Recent advances and trends in science show us that search and recommendation will be transformed in the coming years. One…
Vidéo
Keynote: Learning from observation: Small-data approach to human common sense
Speaker: Katsushi Ikeuchi, Sr. Principal Research Manager, Microsoft Research Redmond Learning-from-Observation (LfO), a robot-teaching paradigm, aims to build a robot system that understands what humans do through a small number of human observations and map…
Vidéo
Research talks: Few-shot and zero-shot visual learning and reasoning
Humans learn, infer, and reason by leveraging prior knowledge without necessarily observing a large number of examples. Visual learning and reasoning technologies, such as few-shot and zero-shot learning, aim to enable human-like learning and reasoning…
